How to Track Pathology Changes Over Time

The short answer

To track pathology changes over time, collect each test under the same conditions, record results in one place with dates and units, and judge change against the marker's normal day-to-day variation.

A change counts when it is larger than that variation, runs in the same direction across several tests and moves with related markers. Anything smaller may be noise.

Why one result is not enough

Two tests on the same healthy person, a week apart, will not give the same numbers. Two things cause this.

Biological variation. Each marker fluctuates around a personal set point. Some, such as sodium, vary little. Others, such as triglycerides and CRP, vary a lot.
Analytical variation. Every laboratory method has a small measurement imprecision.

Laboratory scientists combine the two into a reference change value, the smallest difference between two results that is unlikely to be chance (Fraser, 2011). Published estimates of biological variation are collected in the EFLM Biological Variation Database.

How to make results comparable

A trend is only as good as the consistency of the collections behind it. Ask clients to keep these the same each time.

Same laboratory, where possible
Same time of day, ideally morning
Same fasting status
No hard exercise in the day or two before
Not during or just after an acute illness
Same cycle phase for hormone tests
Supplements and medications recorded, including biotin
Normal hydration

Record any exception beside the result. A result with a known exception is still useful; a result with an unknown one can mislead.

What counts as a real change

A change is more likely to be real when it passes four checks.

It is bigger than the marker's usual variation. As an illustration, a marker with 20% within-person variation and 5% analytical imprecision has a reference change value of about 57% at 95% confidence. A marker with 1% of each has one of about 4%. The same percentage shift means very different things for the two.
It keeps going. Three or more results moving the same way carry more weight than a single step.
Related markers move with it. A fall in ferritin with a fall in transferrin saturation is more convincing than either alone.
It fits the case. The change lines up with symptoms, an intervention or a known event.

Stability is a finding too. A marker that holds steady after an intervention tells you something, as does one that drifts slowly inside its range.

When to retest. Match the interval to the biology. HbA1c reflects roughly the previous two to three months, so retesting sooner adds little. Markers that respond quickly can be rechecked earlier.

Setting up tracking in practice

You need one record per client that holds every result with its date, unit, laboratory and collection notes.

In a spreadsheet. Put markers in rows and test dates in columns. Add a row for fasting status, time of collection and laboratory. Chart the markers you are following. This works, but each new report means retyping values and rebuilding charts.

In a platform. Beyond Normal builds trend charts automatically once a client has more than one test. You upload the pathology PDF, the values are extracted, and each marker is plotted across visits against both its conventional and functional interval. Uploads are unlimited, so you can add a client's older reports to build a longer baseline.

For other tools that track results, see the best online platforms for practitioners to interpret blood tests.

Frequently asked questions

How often should blood tests be repeated to track change?

It depends on the marker and the intervention. HbA1c reflects roughly two to three months, so that is a sensible minimum gap for it. Retest other markers on a timeframe that matches how quickly they respond.

How do I know if a change in a blood result is real?

Check whether it exceeds the marker's normal day-to-day variation, continues across several tests, moves with related markers and fits the clinical picture.

Can I compare results from different laboratories?

With care. Methods and reference intervals can differ between laboratories, so confirm the units and intervals match before comparing. Using the same laboratory each time is more reliable.

How many results do I need to see a trend?

Two results show a difference. Three or more, collected under the same conditions, show a direction.

What is a reference change value?

It is the smallest difference between two results from the same person that is unlikely to be explained by biological and analytical variation alone.

Sources

The citation was checked against PubMed on 4 October 2026.

Fraser CG. Reference change values. Clin Chem Lab Med. 2011;50(5):807-812. doi:10.1515/CCLM.2011.733
EFLM Biological Variation Database