Most lab reports invite a very simple reading: in range is good, out of range is bad.
That is understandable. The colors, arrows, and bolded flags are designed to make a long report easier to scan. But a reference range is not a diagnosis, a treatment target, or a definition of your personal baseline. It is a statistical comparison point.
That distinction matters because a number can be unusual for a population without establishing why it is unusual. And a number can be common in a population while still representing a meaningful change for one person.
What a reference range actually measures
For many routine tests, a reference range is built from results in a selected group of people considered healthy according to specific criteria. Laboratories use statistical methods to identify the central portion of that group’s results, often the middle 95%.
This means that some healthy people will fall outside the range by design. It also means that some people with a health condition will have results inside the range. A reference range is useful context. It is not a clean dividing line between health and disease.
The details also matter. A range may be specific to an age group, sex category, stage of life, specimen type, test method, or laboratory instrument. Two laboratories can report slightly different ranges for the same marker, particularly when methods are not fully harmonized.
A recent review in laboratory medicine revisited how reference intervals are established, including both direct studies of selected reference groups and indirect approaches that use large sets of routine laboratory data. The practical point is simple: the range on a report is constructed. It is not a number handed down by biology.
Population normal is not personal normal
Think of a reference range as a map of where many people’s results fall. It does not automatically show where your results tend to fall when you are well, when your conditions of collection are similar, and when the same method is used.
Some markers vary more between people than within the same person. In those cases, a population range can be wide enough to obscure a meaningful shift from an individual’s usual result. Other markers fluctuate considerably within a person because of time of day, recent food intake, activity, sleep, menstrual-cycle timing, illness, hydration, collection conditions, or ordinary biological variation.
That is why two results that are both marked “normal” are not automatically interchangeable. It is also why a small change, even one that crosses a report’s boundary, is not automatically evidence of a new problem.
Laboratory medicine has a term for this question: the reference change value. It is an estimate of how large a difference between two results needs to be before it is less likely to be explained by the combination of ordinary within-person variation and measurement imprecision. It is a useful concept, but it is not a universal shortcut. The relevant amount of change differs by analyte, laboratory performance, and testing circumstances.
What a flagged result does and does not mean
A flag means the reported value fell outside that laboratory’s stated reference interval. That is all the flag can establish on its own.
It does not establish a diagnosis. It does not explain symptoms. It does not reveal cause. And it does not tell us whether a difference is clinically important.
Some tests use a reference interval. Others are interpreted against a clinical decision limit or diagnostic threshold. These are not the same thing.
A population reference interval describes the distribution of results in a reference group. A clinical decision limit is intended for a particular decision and is ideally connected to outcomes, risk, or diagnostic performance. A treatment target is something else again. And the “optimal” ranges sometimes presented online may reflect professional guidelines, expert convention, laboratory convention, or functional-medicine practice. Those categories should not be treated as interchangeable.
Calling a result “suboptimal” can sound more precise than it is. The useful question is: What outcome data or guideline supports that particular target, for this marker, in this population, using this method?
Why context changes interpretation
A number arrives at the end of a chain of events. The chain includes the person being tested, the timing of the sample, the collection process, handling of the specimen, the assay, and the reference data used for comparison.
Take cortisol as one example. A cortisol result is not simply a measurement of how stressed someone is. Its interpretation depends heavily on the time of collection, the type of sample, the question being asked, sleep timing, acute illness, medications, and the assay itself. The same principle applies broadly. A test may measure something accurately while a broad conclusion drawn from it remains uncertain.
More testing does not remove this problem. It can add useful context, but it can also create more borderline findings, more false positives, and more apparent patterns that do not hold up when collection conditions or repeat measurements are considered.
A better way to read the report
Rather than asking only whether a result is in range, it can be more informative to ask a few narrower questions:
- What exactly did this test measure?
- Is this a population reference interval, a clinical decision limit, or a proposed target?
- Was the sample collected under conditions that could reasonably affect the result?
- Is there a prior result from the same laboratory and method for comparison?
- Would this size of difference exceed ordinary biological and analytical variation?
- What conclusion does this result support, and what conclusion does it not support?
The goal is not to distrust laboratory testing. It is to use laboratory information for what it is good at: adding evidence to a larger question.
A reference range can tell us where a result sits relative to a defined group. It cannot, by itself, tell us what is normal for one individual, what caused a difference, or what should happen next. Those are separate questions, and they deserve separate evidence.
Written for Dr. Teralyn Sell, PhD
Psychology · Brain Health · Human Behavior