Blood tests people order for themselves

Testing without a clinical question is a different activity from testing to answer one, and the difference is where almost all of the trouble comes from.

The statistics of broad testing

Reference ranges are typically constructed to include the central ninety-five per cent of a healthy population.

Which means each individual test has roughly a one in twenty chance of returning outside the range in a well person.

Order twenty analytes and the probability that at least one comes back flagged approaches certainty.

Consumer panels routinely include thirty, fifty or more.

The result is a near-guaranteed abnormality in a healthy person, which then requires explanation, repetition and sometimes imaging.

The cascade

The recognised harm pattern.

An incidental abnormality prompts a repeat test, which prompts a scan, which finds an unrelated incidental finding, which prompts a specialist referral and sometimes a biopsy.

Each step carries cost, anxiety and a small risk of physical harm.

The literature on incidental findings from imaging describes this well, and studies of incidentalomas — unexpected findings on scans — show that most are benign and that pursuing them produces net harm in many cases.

Which is why clinicians order tests to answer questions rather than to look around.

The specific problems with consumer panels

Optimal ranges. Many services apply narrower ranges than laboratory reference intervals, describing results within the normal range as suboptimal.

These thresholds are frequently not validated and frequently correspond to a product the same company sells.

No clinical context. A result means different things depending on symptoms, medication, timing and history, none of which the panel knows.

Timing errors. Cortisol, testosterone and reproductive hormones all require specific timing, which home collection frequently does not control.

Interference. Biotin supplements, commonly taken for hair and nails, interfere with many immunoassays and have produced misleading thyroid and cardiac results.

Follow-up. The abnormal result arrives with no one responsible for acting on it, which transfers the work and the anxiety to the person and their general practitioner.

Where testing does make sense

A useful distinction.

Testing to investigate symptoms, guided by what the symptoms suggest.

Testing to monitor a known condition or treatment.

Population screening programmes, which are designed with evidence about whether earlier detection improves outcomes, and which are the product of considerable analysis rather than availability.

Testing to assess risk where the result would change management — cholesterol and glucose in someone whose treatment decision depends on the number.

And testing where a specific exposure or family history raises a specific question.

The question that distinguishes them all: what will I do differently depending on the answer?

If the answer is nothing, the test is not useful.

Screening that is worth attending

Worth stating, since people who buy panels frequently skip these.

National programmes for cervical, breast and bowel cancer, which have evidence of mortality benefit and are free in many systems.

Blood pressure, which is the most valuable single measurement in preventive medicine and is nearly free.

Cardiovascular risk assessment at the ages recommended locally.

Diabetes screening where risk factors are present.

Sexual health screening according to activity.

And vaccination, which is prevention rather than detection and outperforms almost everything else available.

What to do with a result you already have

Do not panic at a marginally abnormal value, which is common and usually means nothing.

Take it to a clinician with the context: what you were doing, what you take, why you tested.

Expect that many will be repeated rather than acted upon.

Ask whether the result changes anything.

And be aware that presenting a large panel to a general practitioner creates work that the system has not funded, which is not a reason to withhold it and is a reason to be realistic about the response.

The underlying point

The appeal of these services is understandable: they offer agency in systems where getting an appointment is difficult and being taken seriously is not guaranteed.

That is a real problem, and broad untargeted testing is not its solution — it more often produces a new set of things to worry about while the original question remains unanswered.

General information only, not medical advice. Discuss test results with a qualified clinician who knows your history.