Understanding Hormone Laboratory Results: Why Context Matters

Why context matters

Laboratory testing is genuinely valuable — it is one of the best objective tools available for evaluating hormone status, and nothing in this article is intended to suggest otherwise. The point of this article is narrower and more constructive: a single hormone value, viewed without its surrounding context, can be misread in either direction — flagged as abnormal when it reflects normal variation for that person's situation, or read as reassuring when the context (route, timing, symptoms) suggests it needs a closer look. Good laboratory interpretation uses the number and the context together.

Reference ranges

A reference range is a statistical summary of results from a reference population — commonly the range that captures the central 95% of results in that population — not a hard biological boundary between "normal" and "abnormal" for every individual. Where a specific patient's optimal range falls within, at the edge of, or technically outside the population reference range is a clinical judgment, informed by that patient's history and symptoms, not a fact the number alone can establish.

Age, sex and menopausal status

Many hormone reference ranges are stratified by age, sex, and (for reproductive hormones) menopausal or cycle status, because normal physiologic levels genuinely differ across these categories. A result compared against the wrong stratum — for instance, a perimenopausal estradiol level assessed against a premenopausal range without accounting for the transitional, fluctuating nature of perimenopause — can be misleading regardless of how accurate the assay itself is.

Menstrual-cycle timing

For patients who are still cycling, hormone levels — particularly estradiol, progesterone, LH and FSH — vary substantially and predictably across the menstrual cycle. Estradiol rises through the follicular phase as the dominant follicle grows, surges further immediately before ovulation, and falls sharply afterward; progesterone rises sharply in the luteal phase as the corpus luteum forms and predominantly secretes it.^[2] A progesterone level drawn in the follicular phase and one drawn in the luteal phase are expected to differ dramatically as a matter of normal physiology, not as a sign of a problem. Cycle-day context is essential for interpreting these results correctly.

Medication, formulation and route

As detailed in our dedicated article on route of administration, the same hormone at the same dose can produce different absorption curves, peak levels, and metabolite profiles depending on whether it's taken orally, sublingually, topically, or vaginally. A lab value should always be interpreted with the specific formulation and route in mind, not compared against a route-agnostic generic reference range.

Dose timing and sample collection

For routes with a rapid rise and fall in blood levels — sublingual dosing is a good example — the timing of a blood draw relative to the last dose can dramatically change the result: a sample drawn shortly after dosing may capture a transient peak, while one drawn many hours later may reflect a trough. Neither is inherently "correct" — what matters is that the timing is known, consistent between tests where possible, and interpreted with that timing in mind.

Sex hormone-binding globulin (SHBG)

In plasma, testosterone circulates bound primarily to SHBG and albumin, and only the small unbound, "free" fraction is generally considered biologically active at the tissue level.^[1] SHBG is influenced by multiple factors: oral estrogen-containing therapy (including oral contraceptives) raises SHBG, and SHBG is also closely tied to metabolic status, functioning as a strong predictor of insulin resistance.^[1] A total hormone level can look stable while the free, active fraction shifts if SHBG changes. This is why free hormone measurements — either measured directly or calculated from total hormone, SHBG and albumin — are used alongside total levels for some hormones, particularly testosterone, though the Endocrine Society notes there is no single, universally agreed-upon standard for that calculation.^[1]

Serum, saliva and urine testing, at a high level

Serum (blood) testing is the most widely used and validated method for most hormone panels in conventional clinical practice, with established reference ranges and extensive outcome-data correlation. Salivary hormone testing is used by some compounding-oriented practitioners to help guide dosing, but a 2021 peer-reviewed review identified a specific scientific problem with that approach: salivary hormone levels are commonly assumed to be equivalent to the free (non-protein-bound) hormone fraction in blood, and the authors describe that assumption as a misconception rather than an established equivalence, while noting that compounding pharmacies "frequently use saliva testing" on that basis.^[3] For transdermal testosterone monitoring specifically — the most directly studied case — a 2026 review concluded that serum "is the standard method for monitoring testosterone replacement therapy," and that "saliva and DBS [dried blood spot] T testing lack sufficient clinical correlation and should not be used for" transdermal monitoring, since salivary testosterone measured during transdermal therapy "can yield misleading, erroneously high results." The same review found urine testing "tend[s] to parallel serum" testosterone responses but is generally less reliable than serum, with additional variability introduced by genetic differences in hormone metabolism.^[4] This evidence is strongest for transdermal testosterone specifically, and should not be read as a universal rule for every hormone and every route — but the pattern in what has been directly studied favors serum as the default, most validated method, with salivary testing in particular not supported for monitoring hormone therapy dosing.

None of these methods is inherently invalid — each measures something real. The practical question for a patient and prescriber is which method has been validated for the specific clinical question being asked, and whether the reference ranges being used were actually derived from that same testing method and population.

Assay variability

Different laboratories may use different assay platforms (immunoassay vs. mass spectrometry-based methods, for example) with different sensitivity, specificity, and precision — particularly at the low end of a range, as discussed in our testosterone-in-women article. For testosterone specifically, the Endocrine Society has cautioned that direct immunoassays for free testosterone are "seriously flawed and cannot be recommended," regardless of platform, and that consistent calibration of an assay against an agreed-upon standard matters as much as the assay method itself, since different methods "do not necessarily yield identical answers."^[1] This means results from different labs are not always perfectly comparable, and a meaningful change over time is most reliably assessed using the same lab and method when possible.

Symptoms and trends over time

A single lab value is a snapshot; a series of consistently-timed values over time is a trend, and trends are generally more informative than any single point, particularly for hormones with significant natural or route-dependent variability. Symptoms matter clinically, but symptoms alone do not establish a hormone deficiency, and a single lab value should not by itself dictate treatment. As one illustration: the Endocrine Society's clinical practice guideline on androgen therapy in women found that baseline endogenous testosterone levels did not predict which women would respond to treatment for low sexual desire^[1] — a reminder that neither a symptom nor a lab value in isolation reliably answers a clinical question on its own. Symptom tracking alongside lab trends — rather than either in isolation — gives a prescriber the fullest picture for adjusting a treatment plan.