Skip to content
Blood Test Tracker

Cortisol, Late-Night Saliva

Cortisol in a saliva sample taken near bedtime, when levels are normally at their lowest. It shows whether the daily cortisol rhythm dips as expected at night.

AI-verified against cited sources. Awaiting expert reviewAlso called LNSCLOINC 2142-8

Check your result

Sex (optional)

Type your result to see where it falls.

Nothing you type here leaves your browser.

What it is

Cortisol is a hormone made by the adrenal glands. It follows a daily rhythm: it peaks soon after you wake and falls to its lowest point around bedtime. Cortisol in saliva reflects the free, active part of the cortisol in your blood. A saliva sample taken at night is a simple way to see whether that nighttime dip is happening. You collect the sample yourself at home with a cotton swab, usually between about 11 p.m. and midnight or at your usual bedtime, and often on two different evenings.

Why it matters

Clinicians use late-night salivary cortisol as one of several tests when they are looking at whether the body makes too much cortisol over a long time (Cushing syndrome). In that condition the nighttime dip is often lost, so the late-night level stays higher than usual. The test is easy to do and doesn't need a blood draw. Many everyday things can also raise the number, so one result is rarely enough on its own. Clinicians usually repeat it or read it with other tests, such as a 24-hour urine cortisol.

If it’s low

A result at or below your lab's range is the expected pattern. Cortisol is supposed to be lowest around bedtime, so a low late-night value is not a concern on its own. This test is not designed to show whether the body makes too little cortisol. Mayo notes that its midnight salivary assay can't be used to look at low cortisol, such as Addison disease, because of limited sensitivity.

Common causes

  • The normal nighttime low point of the daily cortisol rhythm

What can help

  • If you have ongoing weakness, tiredness, dizziness or weight loss, talk with a clinician. Other tests, such as a morning blood cortisol and ACTH, are the ones used to look at low cortisol

If it’s high

A late-night result above your lab's range means cortisol did not fall as low as expected at night. Most often there is an everyday explanation, such as a collection mistake, a late or irregular bedtime, shift work, stress, or contamination from a steroid cream or tablet. Conditions linked to higher nighttime cortisol include depression, alcohol use disorder, obesity and polycystic ovary syndrome. Less often, a high result is linked to Cushing syndrome. One result can't tell these apart, so clinicians usually repeat the test on another night, and they may add a different type of test before drawing any conclusion.

Common causes

  • A collection problem, such as eating, brushing teeth, mouthwash, alcohol, tobacco or bleeding gums before the sample
  • Contamination from hydrocortisone or other steroid creams, ointments, oral gels or tablets
  • A bedtime far from the time on the kit, shift work, poor sleep or recent travel across time zones
  • Physical or emotional stress, illness or a hospital stay
  • Depression, alcohol use disorder, obesity or polycystic ovary syndrome
  • Pregnancy
  • Cushing syndrome (uncommon)

What can help

  • Follow the collection steps and the timing on your kit exactly
  • Tell your clinician about any cream, gel, tablet or injection that contains a steroid, and about shift work or an unusual sleep schedule
  • Ask whether a repeat sample on another night, or a different kind of cortisol test, would add useful information

Usual ranges

What labs print as the reference range, by who was tested. Ranges differ a little between labs, so the range on your own report is the one your result is judged against.

For adults: up to 0.09 µg/dL. Source: Quest Diagnostics, Labcorp (Endocrine Sciences), Mayo Clinic Laboratories, ARUP Laboratories, Michigan Medicine Laboratories (MLabs), Journal of Clinical Endocrinology & Metabolism, The Lancet Diabetes & Endocrinology, Journal of Endocrinological Investigation.

What can affect your result

  • Timing: the sample has to be taken at the time on your kit, close to your usual bedtime
  • Sleep schedule: night-shift work, a bedtime long after midnight or travel across time zones can change the nighttime low point
  • Stress right before the sample, illness, or a hospital stay
  • Mouth contamination: food, drink, brushing, mouthwash, lip products, tobacco and bleeding gums
  • Steroid creams, gels, tablets and injections
  • Depression, alcohol use disorder, obesity and polycystic ovary syndrome
  • Age and health conditions: in one study of men aged 60 and older, 20% had at least one raised late-night result, and 40% of those with diabetes and high blood pressure did (Endocrine Society 2008)
  • Assay method: LC-MS/MS and immunoassay results and cutoffs differ
  • Collection steps matter: ARUP asks for no food for 60 minutes, no alcohol for 12 hours, no tooth brushing right before (bleeding gums can add blood and raise the result), no mouthwash and no lip products. Rinse your mouth with water 10 minutes before the sample.
  • Hydrocortisone creams, ointments, oral gels and tablets, and other steroid products, can contaminate the sample or raise cortisol. Tell your clinician about any you use.
  • Smoking has been linked to higher late-night results, and licorice or chewing tobacco may raise them (Endocrine Society 2008).
  • Shift work, a bedtime long after midnight, or recent travel across time zones can change the nighttime low point. The test isn't suited to people without a regular day-night cycle.
  • Stress right before the sample, illness or a hospital stay, depression, alcohol use disorder, obesity and polycystic ovary syndrome can raise nighttime cortisol.
  • Some medicines, including steroid medicines and anti-seizure drugs, can raise cortisol or flatten its daily rhythm (Mayo).
  • LC-MS/MS and immunoassays give different numbers, so compare results from the same method.

Late evening, as your kit says. Mayo prefers 11 p.m. to midnight, ARUP recommends 11 p.m. to 1 a.m., and Quest's range covers 10 p.m. to 1 a.m. The 2021 consensus notes that sampling at your usual bedtime, rather than at midnight, could reduce false positives because the cortisol low point follows the start of sleep. A single sample is a snapshot of that night. In cyclic Cushing syndrome, cortisol can be high in some weeks and normal in others, which is one reason for repeat samples.

When to talk to a clinician

There is no emergency threshold for this test, and a result above your lab's range is not urgent on its own. Talk with the clinician who ordered it about next steps. Michigan Medicine treats a late-night result of 100 ng/dL (0.10 µg/dL) or higher as one that should be followed by a second test, and your own lab's cutoff may differ. Let a clinician know about symptoms that can go with high cortisol, such as weight gain around the middle, easy bruising, wide purple stretch marks, new high blood pressure or high blood sugar. Seek urgent care for severe weakness, fainting, severe belly pain, vomiting or confusion, especially if you take or recently stopped steroid medicines. Call 911 or your local emergency number if you feel very unwell.

Questions you could ask

  • How many nights should I collect a sample, and at what time?
  • Should repeat tests use the same lab and method?
  • Could my collection steps, bedtime, sleep schedule, or a cream or medicine I use have affected this result?
  • Should we repeat the test on another night?
  • Would a different kind of cortisol test, such as a 24-hour urine collection, add useful information?
  • Is a low late-night result what you would expect?

Questions people ask

Why use saliva instead of blood?

Saliva holds the free, active part of cortisol, and you can collect it at home at the right time of night. A blood draw around midnight is harder to arrange. Your clinician will tell you which type of cortisol test fits your situation.

Does a result above my lab's range mean I have Cushing syndrome?

No. A raised late-night result has many everyday explanations, including collection mistakes, a shifted sleep schedule, stress and steroid creams. That is why clinicians repeat the test, often on another night, and sometimes add a different type of cortisol test before drawing any conclusion.

How do I convert the units?

Multiply µg/dL by 10 to get ng/mL, by 1,000 to get ng/dL, or by 27.6 to get nmol/L. For example, 0.09 µg/dL is 90 ng/dL, 0.9 ng/mL or about 2.5 nmol/L.

Show the science and sources

Units

  • 1 µg/dL = 1 µg/dL
  • 1 µg/dL = 27.59 nmol/L
  • 1 µg/dL = 10 ng/mL
  • 1 µg/dL = 1000 ng/dL

Open the Late-night salivary cortisol unit converter

For clinicians and researchers

Salivary cortisol reflects the free (unbound) fraction in blood, is in equilibrium with it within minutes, and isn't affected by the rate of saliva production (Endocrine Society 2008). The test relies on loss of the late-night cortisol nadir in Cushing syndrome. Across assays and cutoffs, published studies report 92-100% sensitivity and 93-100% specificity, with accuracy similar to urinary free cortisol (Endocrine Society 2008). The 2008 guideline recommends two measurements, and notes that most clinicians ask for samples on two separate evenings between 11 p.m. and midnight. It cited normal bedtime values under 145 ng/dL (4 nmol/L) for the ELISA and LC-MS/MS assays then in use. Cutoffs are assay-specific. Current LC-MS/MS laboratory limits are about 0.09-0.10 µg/dL (90-100 ng/dL, 2.5-2.8 nmol/L): Quest 0.09, Labcorp 0.09, Mayo, ARUP and Michigan Medicine 0.10. A 2020 study of 100 healthy volunteers found a late-night LC-MS/MS interval of 0.17-0.97 µg/L and a positive bias of 51.8% for the immunoassay over LC-MS/MS. The 2021 consensus advises two or three samples, bedtime rather than midnight sampling to reduce false positives, and notes that mass spectrometry has better sensitivity than immunoassay at the cost of lower specificity. Mass spectrometry that also measures cortisone, as Mayo's does, can help flag contamination from topical hydrocortisone. Salivary glands convert some cortisol to inactive cortisone through 11-beta-HSD2, so licorice (glycyrrhizic acid) and chewing tobacco may, in theory, raise results. The 2021 consensus notes lower specificity in patients with adrenal tumors and starts with a dexamethasone suppression test when an adrenal tumor is suspected, and the test should not be used in people without a normal day-night cycle. For monitoring after pituitary surgery, the 2021 consensus lists a late-night cutoff of 0.27 µg/dL (7.5 nmol/L) with 75-90% sensitivity and 93-95% specificity, and suggests yearly testing. Mayo suggests confirming a raised result with a repeat late-night sample, midnight serum cortisol, 24-hour urinary free cortisol or an overnight dexamethasone suppression test. A Salivette sample correlates better with serum cortisol than passive drool, though it reads lower, and samples are stable for weeks at room or refrigerator temperature (Endocrine Society 2008).

LOINC 2142-8 (primary), 58642-0, 51844-9, 58643-8. Specimen: saliva.

Sources

  1. Cortisol, LC/MS, Saliva (test code 19897), Quest Diagnostics
  2. Salivary Cortisol, LC/MS-MS (test 500179), Labcorp (Endocrine Sciences)
  3. SALCT: Cortisol, Saliva, Mayo Clinic Laboratories
  4. Cortisol by LC-MS/MS, Salivary (ARUP test 3016866), ARUP Laboratories
  5. Cortisol, Saliva (SCOR), Michigan Medicine Laboratories (MLabs)
  6. The Diagnosis of Cushing's Syndrome: An Endocrine Society Clinical Practice Guideline, Journal of Clinical Endocrinology & Metabolism (2008)
  7. Consensus on Diagnosis and Management of Cushing's Disease: A Guideline Update, The Lancet Diabetes & Endocrinology (2021)
  8. Reference ranges of late-night salivary cortisol and cortisone measured by LC-MS/MS and accuracy for the diagnosis of Cushing's syndrome, Journal of Endocrinological Investigation (2020)
  9. Cortisol (PubChem CID 5754), PubChem (NIH)

AI-verified against cited sources. Awaiting expert review. Last reviewed 2026-10-03. How we build and check these pages.

Related markers