How to Test Cortisol Levels: Types, Timing, and Results

Cortisol can be measured in blood, saliva, urine, or hair, and each sample type captures a different window of the hormone’s activity. Blood draws and saliva swabs reflect what your body is doing in real time, a 24-hour urine collection averages an entire day’s output, and a hair sample can reach back months. Which test you need depends on what your doctor is looking for, and the timing of any single sample matters enormously because cortisol follows a steep daily rhythm that can shift a “normal” reading to an “abnormal” one in a matter of hours.

Why the Clock Matters More Than You Think

Cortisol is not a steady-state hormone. It follows a pronounced 24-hour cycle driven by the brain’s internal clock, peaking in the early morning and dropping to its lowest point around midnight.¹ That morning surge, called the cortisol awakening response, typically hits its highest point about 30 to 60 minutes after you wake up.² From there, levels decline through the afternoon and evening, reaching a nadir late at night.

This rhythm is so steep that a blood cortisol level drawn at 8 a.m. and one drawn at 4 p.m. can look like they belong to two different people. Standard morning reference ranges for serum cortisol typically fall roughly between 6 and 18 micrograms per deciliter, but an afternoon draw will naturally be much lower. If a lab report does not specify when the blood was taken, the number is almost uninterpretable. For that reason, most routine blood cortisol draws are scheduled first thing in the morning, usually between 7 and 9 a.m., so results can be compared against established reference ranges.

Even small timing errors can distort results in research settings. When scientists measure the cortisol awakening response from saliva, a delay of just four to six minutes between actually waking up and collecting the sample can inflate the reading, while a delay of seven to fifteen minutes can underestimate it.³ In clinical practice the stakes of a few minutes are lower, but the principle holds: when you collect the sample changes what the number means.

Blood Tests and the Free-Versus-Bound Problem

A standard serum cortisol test measures total cortisol in your blood. The catch is that roughly 80% of the cortisol circulating in your bloodstream is bound to a carrier protein called cortisol-binding globulin (CBG), and only the unbound “free” fraction is biologically active.⁴ In a healthy person under normal circumstances, total cortisol tracks free cortisol closely enough for clinical purposes. But anything that changes CBG levels can throw off the relationship.

Pregnancy, estrogen-containing birth control pills, and severe illness are the most common situations where CBG shifts enough to matter. Oral contraceptives raise CBG, which pushes total cortisol readings higher even though the free, active cortisol may be perfectly normal. Severe illness does the opposite: CBG drops during acute inflammation, so total cortisol can look deceptively low while free cortisol remains adequate.⁴ In people who lack CBG entirely due to a rare genetic variant, more than 25% of total cortisol floats free instead of the usual 6% or less, and the rest binds loosely to albumin.⁵

When the clinical picture suggests CBG might be abnormal, some labs can calculate a free cortisol index (total cortisol divided by CBG) or measure free cortisol directly. These are not routine tests, though. For most people getting a standard blood draw, the total cortisol result is sufficient as long as the doctor knows about medications or conditions that change binding proteins.

Saliva Tests

Salivary cortisol has become one of the most practical ways to measure the hormone, especially for screening purposes. Cortisol in saliva reflects the free fraction in blood, sidestepping the binding-protein issue entirely. Studies have consistently shown a strong correlation between salivary cortisol and serum free cortisol across a wide range of conditions, including both healthy people and those with adrenal disorders.⁶ That correlation holds across different assay methods as well.⁷ Even in critically ill patients with septic shock, salivary cortisol tracked free serum cortisol closely.⁸

The collection itself is simple: you chew on or hold a cotton swab in your mouth for a minute or two, then seal it in a tube. No needles, no lab visit. That convenience makes saliva especially useful for late-night samples, which are the gold standard for screening Cushing’s syndrome. People with Cushing’s lose the normal evening drop in cortisol, so a bedtime salivary cortisol measurement that comes back elevated is a strong signal. Studies have found that a late-night salivary cortisol measurement correctly identified about 92 to 93% of patients with Cushing’s syndrome while excluding those without it.⁹¹⁰

A few practical details can affect salivary cortisol readings. Eating a meal bumps cortisol modestly, and exercise can push levels up substantially, by roughly 80% in one study of healthy individuals.¹¹ Most protocols ask you to avoid eating, drinking anything other than water, and brushing your teeth for at least 30 minutes before collecting the sample. Blood contamination from bleeding gums can also skew results.

24-Hour Urine Collection

When a doctor needs to know your total daily cortisol output rather than a snapshot at one time of day, a 24-hour urine free cortisol test is the standard approach. You collect every drop of urine over an entire day-night cycle into a single container, and the lab measures the unbound cortisol that the kidneys filtered out. Because the collection spans a full circadian cycle, it averages out the peaks and troughs and gives an integrated picture of cortisol production that is not affected by CBG levels.¹²

The trade-off is that the collection is cumbersome. You have to carry a collection jug around for a full day, remember to use it every time, and keep it refrigerated. Incomplete collections are the biggest source of error; labs typically measure urinary creatinine alongside cortisol to check whether the collection looks complete. If you accidentally missed a void or two, creatinine will be suspiciously low, and the lab or your doctor should flag that the cortisol number is unreliable. The 24-hour urine test is one of the recommended first-line screening tools for Cushing’s syndrome, often paired with a late-night salivary cortisol or a suppression test to increase diagnostic confidence.

Hair Cortisol for Long-Term Patterns

Blood, saliva, and urine all capture cortisol over hours at most. Hair cortisol does something fundamentally different: it measures cumulative cortisol exposure over months. As hair grows at a fairly steady rate of about one centimeter per month, cortisol circulating in the blood gets incorporated into the growing strand. A lab can cut a small section of hair near the scalp, divide it into segments, and reconstruct a month-by-month cortisol timeline.¹³

This makes hair cortisol useful in a way that no other test is. Researchers studying chronic stress, for instance, can use a single hair sample to look back at cumulative stress-hormone exposure across three or more months, much like hemoglobin A1c gives a retrospective window on blood sugar control.¹⁴ The test is not commonly used in routine clinical practice yet, but it is gaining traction in research on chronic stress and metabolic disease, and it has been used to evaluate whether stress-reduction interventions are actually changing cortisol levels over time.¹⁵

Hair cortisol does have limitations. Hair treatments like bleaching and dyeing can degrade cortisol in the strand. People with very short hair may not have a long enough sample for multi-month analysis. And it cannot tell you what cortisol is doing right now, which matters when the clinical question is about an acute problem like adrenal crisis.

Dynamic Tests Your Doctor Might Order

Sometimes a single cortisol measurement is not enough to figure out whether your adrenal glands are working properly. Dynamic tests deliberately push the system in one direction and then measure how it responds.

The dexamethasone suppression test is the most common. You take a low dose of dexamethasone, a synthetic steroid, at 11 p.m. the night before a morning blood draw.¹⁶ Dexamethasone tells the brain that there is already plenty of steroid in the system, so a healthy pituitary gland responds by dialing down its signal to the adrenals. If your morning cortisol suppresses below a standard cutoff, the feedback loop is working normally. If cortisol stays elevated despite the dexamethasone, it raises suspicion for Cushing’s syndrome or another source of excess cortisol. Newer lab methods can simultaneously measure dexamethasone levels in the blood to confirm that the drug was actually absorbed properly, which helps avoid misinterpreting the result.¹⁷

The ACTH stimulation test works in the opposite direction. A synthetic version of ACTH, the hormone the pituitary uses to tell the adrenals to produce cortisol, is injected, and blood cortisol is measured 30 to 60 minutes later. A healthy adrenal gland should respond with a brisk rise in cortisol. If it does not, the test suggests adrenal insufficiency. A meta-analysis of studies using this test found that both high-dose and low-dose versions had similar accuracy. Both were good at confirming adrenal insufficiency when the result was abnormal, but neither was reliable at ruling it out when the result came back normal.¹⁸ In practice, this means a normal ACTH stimulation test does not always guarantee your adrenals are fine, especially if symptoms are suggestive.

What Can Throw Off Your Results

Oral contraceptives are one of the most common causes of confusing cortisol numbers. Estrogen-containing pills raise CBG, which inflates total serum cortisol. More than half of women on oral contraceptives may get a falsely abnormal result on a dexamethasone suppression test.¹⁹ This is well known to endocrinologists, but it can catch general practitioners off guard. If you are taking estrogen-containing contraception and your cortisol results come back high, the medication is the first thing to consider before pursuing further workup. Salivary cortisol is less affected because it measures free cortisol, but the issue is not completely eliminated.

Shift work creates a different kind of problem. Night-shift workers tend to show a blunted cortisol rhythm: their morning rise is flatter and their evening levels are higher than those of day workers, producing a compressed daily pattern.²⁰ Their circadian rhythm still exists but it is dampened, even on days off.²¹ This means standard reference ranges calibrated to daytime workers may not fit. If you work nights and need cortisol tested, make sure your doctor knows your schedule, because “morning” cortisol for you might functionally be “evening” cortisol by your body’s clock.

Other everyday factors that bump salivary cortisol include exercise, which in one study raised levels by about 80%, and meals, which caused a modest increase of around 10%.¹¹ Acute stress, caffeine, alcohol, and certain medications including inhaled corticosteroids and topical steroid creams can also move the needle. The general advice is to follow the sample-collection instructions closely and avoid vigorous activity or eating in the window before collection.

The Lab Method Matters More Than You Might Expect

Not all cortisol assays produce the same numbers from the same sample. The two broad categories of measurement technology are immunoassays, which use antibodies to detect cortisol, and liquid chromatography-tandem mass spectrometry (LC-MS/MS), which separates and identifies molecules by their chemical structure. Immunoassays are faster and cheaper and are what most hospital and commercial labs use. LC-MS/MS is more precise and more specific but requires expensive equipment and skilled operators.

The practical issue is that immunoassays can cross-react with other steroid molecules in the sample, producing readings that are systematically higher than what LC-MS/MS reports. One study of salivary cortisol in children found that immunoassay values ran about 2.4 times higher than mass spectrometry values, a difference driven partly by cross-reactivity with cortisone at low concentrations.²² Broader comparisons of commonly used immunoassays against LC-MS/MS have found nonlinear relationships, meaning the degree of disagreement changes depending on the cortisol level.²³ For urinary free cortisol, a recent comparison of four commercial immunoassays against LC-MS/MS found good overall correlation, but all four immunoassays read higher, with the average bias ranging from about 11% to 51% depending on the platform.²⁴

For patients, the takeaway is that reference ranges are method-specific. A “normal” cortisol value on one lab’s immunoassay platform is not directly comparable to a value from a different lab using mass spectrometry. If you are tracking cortisol over time, staying with the same lab and the same assay method produces more reliable comparisons than switching between providers. Mass spectrometry also has a diagnostic advantage in that it can measure cortisol alongside other steroids like cortisone and dexamethasone in a single run, which helps doctors verify whether a suppression test was properly performed or whether a patient has been taking exogenous steroids.²⁵

Pregnancy and Cortisol

Pregnancy naturally and dramatically raises cortisol. The placenta produces corticotropin-releasing hormone, which ramps up the entire hormonal cascade, and estrogen-driven increases in CBG push total serum cortisol even higher. In one cohort study, mean maternal serum cortisol rose from about 10.3 micrograms per deciliter at 24 weeks of pregnancy to about 12.2 micrograms per deciliter by 29 weeks.²⁶ Another study tracking cortisol from early pregnancy onward found total cortisol roughly increased from about 390 nanomoles per liter at the fifth week to about 589 nanomoles per liter by the twentieth week.²⁷

These elevations are physiologically normal and do not indicate Cushing’s syndrome. But they make diagnosing genuine cortisol disorders during pregnancy extremely difficult. Standard cutoffs for the dexamethasone suppression test and serum cortisol do not apply. When screening for Cushing’s in a pregnant patient is genuinely needed, endocrinologists typically rely on salivary cortisol, since it better reflects free cortisol, or on a carefully interpreted 24-hour urine collection, though even urine free cortisol rises during pregnancy. The bottom line is that cortisol testing during pregnancy requires specialist interpretation, and results that look alarming on paper often reflect normal physiology.

At-Home Cortisol Kits

A growing number of direct-to-consumer companies sell at-home saliva cortisol kits. You collect saliva samples at specified times of day, mail them to a lab, and get results online. The appeal is obvious: no doctor’s visit, no blood draw, and a window into your stress hormones from the comfort of your living room.

The science behind saliva testing is legitimate, as covered above. The concern is less about the biology and more about the infrastructure around it. A review of ambulatory cortisol-sensing technologies noted that most emerging self-testing devices were still at the proof-of-concept stage and not yet ready for evaluation against clinical diagnostic standards.²⁸ Even for kits that use standard commercial immunoassays, the lack of clinical context is a real limitation. A slightly elevated bedtime cortisol in an at-home kit does not mean you have Cushing’s syndrome any more than a slightly low morning reading means you have adrenal insufficiency. Without a clinician to factor in your medications, sleep schedule, collection compliance, and symptoms, a raw number can generate anxiety out of proportion to its clinical meaning.

If you want to use an at-home kit as a conversation starter with your doctor, that is reasonable. Treat the results as a signal to investigate further rather than a diagnosis in themselves. And pay attention to the practical details that affect accuracy: freeze samples if the kit instructions say to, avoid eating or brushing teeth beforehand, and note the exact time of each collection. Research on sample storage shows that saliva intended for cortisol testing should be frozen within 24 hours and analyzed within about a month for best accuracy.²⁹ Centrifuged samples can be stored refrigerated for up to three months or frozen for at least a year, but leaving saliva at room temperature degrades the cortisol over time.³⁰

How Waking Method Affects the Morning Cortisol Surge

Here is a detail most people never consider: how you wake up can change the size of your morning cortisol spike. In one study, participants who woke to an alarm clock on both workdays and days off showed a cortisol awakening response of about 100%, meaning their cortisol roughly doubled within the first hour. But on days off when they woke spontaneously without an alarm, the response was only about 39%.¹¹ Research on the circadian underpinnings of this response found that the cortisol awakening surge is itself under circadian control, peaking at a phase corresponding to about 3:40 to 3:45 a.m. and essentially absent during afternoon circadian phases.²

For anyone collecting a morning saliva or blood sample, this means that the conditions of waking matter. An alarm-clock jolt on a Monday morning produces a different cortisol curve than a lazy Sunday where you drift awake naturally. If you are doing serial testing on different days, try to keep your waking conditions consistent, and always note on the sample label exactly what time you woke up and how. Researchers studying the cortisol awakening response now routinely use electronic monitoring devices to timestamp the moment of waking and the moment of sample collection, precisely because small timing errors compound.³ You do not need electronic monitoring for a clinical test, but the principle is worth internalizing: cortisol testing rewards precision, and “approximately 8 a.m.” is not as informative as “woke at 7:12, collected at 7:45.”

1
PubMed Central. Cortisol on Circadian Rhythm and Its Effect on Cardiovascular System
2
PubMed Central. The circadian system modulates the cortisol awakening response in humans
3
PubMed. Assessment of the cortisol awakening response: Real-time analysis and curvilinear effects of sample timing inaccuracy
4
PubMed. Free cortisol index is better than serum total cortisol in determining hypothalamic-pituitary-adrenal status in patients undergoing surgery
5
PubMed. Plasma free cortisol fraction reflects levels of functioning corticosteroid-binding globulin
6
PubMed. Salivary cortisol: a better measure of adrenal cortical function than serum cortisol
7
Practical Laboratory Medicine. Comparative analysis of salivary cortisol measurements using different assay methods in relation to serum-free cortisol measurement
8
PubMed Central. Salivary Cortisol Can Replace Free Serum Cortisol Measurements in Patients With Septic Shock
9
PubMed. Late-night salivary cortisol as a screening test for Cushing’s syndrome
10
The Journal of Clinical Endocrinology & Metabolism. Nighttime Salivary Cortisol: A Useful Test for the Diagnosis of Cushing’s Syndrome
11
PubMed. Effects of lifestyle factors on concentrations of salivary cortisol in healthy individuals
12
PubMed Central. Urine free cortisol in the diagnosis of Cushing’s syndrome: is it worth doing and, if so, how?
13
PubMed. Hair cortisol as a biological marker of chronic stress: current status, future directions and unanswered questions
14
PubMed Central. Hair Cortisol Analysis: A Promising Biomarker of HPA Activation in Older Adults
15
PubMed. Hair cortisol, stress exposure, and mental health in humans: a systematic review
16
The Journal of Clinical Endocrinology & Metabolism. Interpretation of Abnormal Dexamethasone Suppression Test is Enhanced With Use of Synchronous Free Cortisol Assessment
17
European Journal of Endocrinology. P18 – A new LC-MS/MS method for the determination of serum cortisol, cortisone, and dexamethasone: utility in the interpretation of the low-dose dexamethasone suppression test
18
The Journal of Clinical Endocrinology & Metabolism. ACTH Stimulation Tests for the Diagnosis of Adrenal Insufficiency: Systematic Review and Meta-Analysis
19
PubMed Central. False-Positive Urinary Free Cortisol with Oral Contraceptive Use
20
PubMed Central. A Systematic Review of Literature on the Association Among Sleep, Cortisol Level and Cardiovascular Health Within the Healthcare Shift Worker Population
21
PubMed Central. Effect of night-shift work on cortisol circadian rhythm and melatonin levels
22
PubMed. Immunoassay or LC-MS/MS for the measurement of salivary cortisol in children?
23
PubMed. Comparison of salivary cortisol as measured by different immunoassays and tandem mass spectrometry
24
PubMed Central. Comparative evaluation of four new immunoassays and LC-MS/MS for the measurement of urinary free cortisol in Cushing’s syndrome diagnosis
25
PubMed. Measurement of cortisol, cortisone, prednisolone, dexamethasone and 11-deoxycortisol with ultra high performance liquid chromatography-tandem mass spectrometry
26
PubMed Central. Normative serum cortisol levels in second and third trimesters and their associated factors: A prospective cohort study from Sri Lanka
27
PubMed. Maternal serum cortisol levels during pregnancy differ by fetal sex
28
Sensing and Bio-Sensing Research. Ambulatory measurement of cortisol: Where do we stand, and which way to follow?
29
PubMed. Measurement of steroid hormones in saliva: Effects of sample storage condition
30
PubMed. Long-term stability of salivary cortisol