Cortisol is the body's principal glucocorticoid and one of the most extensively studied molecules in human biology. It sits at the intersection of stress physiology, circadian regulation, immune modulation, and metabolic health. Yet despite its ubiquity in clinical panels and consumer wellness testing, cortisol remains one of the most easily misread biomarkers available. Interpretation without knowledge of collection timing, binding protein status, assay cross-reactivity, and companion analytes risks systematic misclassification.
This article synthesises four decades of research to map cortisol's biology, its dysregulation in chronic stress and psychiatric illness, and the measurement framework required to extract actionable insight from a cortisol result.
1. The HPA Axis: From Hypothalamic Signal to Adrenal Output
Cortisol is synthesised in the zona fasciculata of the adrenal cortex through a tightly regulated neuroendocrine cascade known as the hypothalamic-pituitary-adrenal (HPA) axis. Corticotropin-releasing hormone (CRH), secreted by parvocellular neurons in the paraventricular nucleus of the hypothalamus, travels via the hypophysial portal system to the anterior pituitary, where it stimulates corticotroph cells to cleave pro-opiomelanocortin (POMC) and release adrenocorticotropic hormone (ACTH). ACTH then activates a cAMP-mediated cascade in the adrenal cortex — driving cholesterol through a series of enzymatic conversions (CYP11A1, CYP17A1, HSD3B2, CYP21A2, CYP11B1) — to produce cortisol.
Daily cortisol production averages 20–30 mg, with acute stress capable of inducing a 10- to 12-fold surge in secretion. Approximately 90–95% of circulating cortisol is protein-bound — roughly 80% to cortisol-binding globulin (CBG) and 10% to albumin — leaving only 5–10% as the biologically active free fraction. This distinction between total and free cortisol is foundational to correct interpretation.
1.1 Glucocorticoid Effects Span Metabolism, Immunity and Cardiovascular Function
Cortisol exerts pleiotropic effects through genomic and non-genomic pathways mediated by the glucocorticoid receptor (GR). Metabolically, cortisol stimulates hepatic gluconeogenesis, promotes protein catabolism in skeletal muscle, induces insulin resistance via impaired GLUT4 translocation, and drives lipolysis with paradoxical central fat redistribution. GR-mediated transcriptional regulation has been estimated to influence up to 20% of the human genome.
Immunologically, cortisol operates on a dose-dependent continuum. Physiological concentrations prime innate immune responses; stress-level concentrations suppress immune activation through transrepression of NF-κB and AP-1 transcription factors. Cortisol suppresses Th1 and Th17 responses, inhibits pro-inflammatory cytokines (IL-1, IL-6, TNF-α), and promotes anti-inflammatory macrophage phenotypes. This immunosuppressive capacity underlies both the therapeutic value of glucocorticoid drugs and the immunological vulnerability associated with chronic stress states.
1.2 Feedback Regulation Through a Dual-Receptor System
Cortisol feedback operates through two receptors with fundamentally different physiological roles. The mineralocorticoid receptor (MR), with its high binding affinity (Kd ~0.5 nM), is substantially occupied at basal cortisol levels and is predominantly expressed in hippocampal and limbic structures. MR provides tonic "proactive" feedback that maintains basal HPA axis tone and sets the circadian rhythm amplitude. The glucocorticoid receptor (GR), with lower affinity (Kd ~5 nM) and ubiquitous expression, becomes occupied primarily during circadian peaks and stress, mediating "reactive" feedback that suppresses CRH and ACTH transcription and facilitates recovery.
Three temporal scales of feedback operate simultaneously: ultra-rapid non-genomic feedback (seconds to minutes), fast genomic feedback (minutes to hours), and long-loop delayed feedback via GR at the hypothalamus and pituitary. Research indicates that MR blockade significantly increases stress-induced cortisol release — confirming MR's critical role in early negative feedback — while GR blockade has no acute dampening effect, consistent with GR's role in delayed sustained termination of the stress response.
2. Circadian Architecture: Pulsatility, the Cortisol Awakening Response, and Clock-Gene Interactions
Cortisol follows one of the most robust circadian rhythms detectable in the human endocrine system. Serum cortisol peaks at the sleep-wake transition (6–8 AM; typical range 10–20 μg/dL or 275–550 nmol/L) and reaches a nadir around midnight (<5 μg/dL or <140 nmol/L). This rhythm is not a continuous wave but is composed of approximately hourly ultradian pulses, with circadian variation arising from amplitude modulation of these pulses rather than frequency changes.
The rhythm is driven by the suprachiasmatic nucleus (SCN) master clock via three mechanisms:
- Circadian modulation of CRH and ACTH secretion amplitude
- Autonomic nervous system regulation of adrenal sensitivity to ACTH via the splanchnic nerve
- The adrenal cortex's own intrinsic peripheral circadian clock
2.1 The Cortisol Awakening Response Is a Distinct Neuroendocrine Event
The cortisol awakening response (CAR) — a 50–160% rise in salivary cortisol within the first 30–45 minutes after morning awakening — is not merely the peak of the circadian rise but a neurobiologically distinct phenomenon. It is regulated by a separate SCN extra-pituitary neural pathway that directly sensitises adrenal ACTH receptors before the ACTH stimulus arrives.
Expert consensus from the International Society of Psychoneuroendocrinology specifies that valid CAR assessment requires: objective verification of awakening time (e.g., actigraphy), samples collected at awakening, +15 min, and +30 min minimum, and at least two sampling days. Critically, an elevated CAR is associated with current major depression and predicts depressive episodes up to 2.5 years in advance, while an attenuated or absent CAR characterises post-traumatic stress disorder and correlates with its severity.
The most current integrative model proposes dual CAR functions: a primary process that mobilises energy resources for anticipated daily demands, and a secondary process that counterregulates adverse prior-day emotional experiences.
2.2 Molecular Clock Components Directly Gate Glucocorticoid Action
The interaction between the circadian clock and glucocorticoid signalling is bidirectional. Cryptochromes CRY1 and CRY2 interact with the GR in a ligand-dependent manner and globally oppose GR transcriptional activation — CRY deficiency approximately doubles the number of dexamethasone-responsive genes. Conversely, glucocorticoids reset peripheral clocks via glucocorticoid response elements (GREs) in clock gene promoters, particularly Per1 and Per2. This makes circulating cortisol the major internal synchroniser of peripheral circadian clocks throughout the body.
Shift work disrupts this entire system. A scoping review of 38 studies found that night shift work significantly disrupts cortisol levels, diurnal rhythm, and the CAR, with irregular schedules producing greater disruption than regular ones. Recovery requires at least two consecutive days off. Chronic cortisol rhythm disruption in shift workers is associated with metabolic syndrome, cardiovascular disease, impaired immune surveillance, and elevated mood disorder risk.
3. Stress, Allostatic Load, and the Transition From Adaptation to Disease
3.1 Acute Stress Is Adaptive; Chronic Stress Remodels the HPA Axis
The acute cortisol response to stress is a well-characterised adaptive mechanism. Standard laboratory stress protocols reliably induce 2- to 4-fold salivary cortisol elevations, mobilising glucose, suppressing non-essential functions, and priming the organism for action. This is physiologically appropriate.
Chronic stress, however, fundamentally remodels HPA axis function along a predictable trajectory. A landmark meta-analysis resolved decades of contradictory literature: cortisol tends to be elevated at stressor onset but declines as stress chronifies, and chronic uncontrollable stress can ultimately produce hypocortisolism — characterised by a high, flat diurnal cortisol profile with loss of normal circadian variation. This cortisol "flattening" has now been replicated across multiple stress paradigms and is associated with elevated inflammatory biomarkers.
Research indicates that persistent high perceived stress is associated with blunted cortisol reactivity to acute challenge — a finding with direct implications for how we interpret single-point cortisol measurements in chronically stressed individuals.
3.2 McEwen's Allostatic Load Framework Connects Cortisol to Systemic Disease
Bruce McEwen's concept of allostatic load — the cumulative physiological "wear and tear" from repeated or chronic stress — provides the theoretical bridge between cortisol dysregulation and disease. McEwen identified three pathological patterns: frequent activation of allostatic systems, failure to shut off the stress response after a challenge, and inadequate response leading to compensatory overactivation of other mediators.
The MacArthur Studies empirically validated this model: in a cohort of elderly adults, a composite allostatic load score incorporating HPA axis, sympathetic, cardiovascular, and metabolic markers predicted 7-year all-cause mortality, cognitive decline, and physical functioning decline — independently of demographics and baseline health status. Cortisol was a component biomarker within this composite.
3.3 Depression Is Associated With Hypercortisolism; Burnout With Hypocortisolism
The relationship between cortisol and psychiatric conditions follows a biphasic pattern that maps onto the allostatic load continuum. The largest meta-analysis to date — encompassing over 670 effect sizes across 361 studies — found that individuals with major depressive disorder show significantly elevated cortisol (d = 0.60) and ACTH (d = 0.28), with stronger effects in older adults, inpatients, and the melancholic subtype. This HPA axis hyperactivity is thought to arise from glucocorticoid receptor resistance that impairs negative feedback — a mechanism consistent with the frequently observed "non-suppression" on the dexamethasone suppression test.
Importantly, atypical depression shows the opposite pattern: hypocortisolism. Clinical burnout patients similarly exhibit functional hypocortisolism — significantly lower urinary free cortisol than healthy controls — a state that may emerge only at the most advanced stages of HPA axis exhaustion following a period of initial hypercortisolism.
3.4 Local Cortisol Amplification Drives Metabolic Syndrome
The link between cortisol and metabolic dysfunction is mediated not primarily by circulating levels but by local cortisol regeneration via 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1). Transgenic mice overexpressing 11β-HSD1 selectively in adipose tissue developed visceral obesity, insulin-resistant diabetes, and hyperlipidaemia — a complete metabolic syndrome phenotype — despite normal circulating glucocorticoid levels. In human obesity, 11β-HSD1 is elevated in adipose tissue, creating localised glucocorticoid excess even when serum cortisol appears normal.
This mechanism explains the striking overlap between Cushing's syndrome features (visceral adiposity, hypertension, dyslipidaemia, impaired glucose tolerance) and metabolic syndrome — and underscores why serum cortisol alone is an insufficient lens for evaluating glucocorticoid-related metabolic risk. Biomarkers such as the Biostarks Metabolic Health panel, which captures multiple facets of metabolic function, complement cortisol assessment in this context.
4. Measurement Methods and Their Caveats
Cortisol can be measured across four biological matrices, each capturing a distinct temporal dimension of HPA axis activity and presenting its own analytical challenges.
4.1 Serum Cortisol: The Clinical Standard With Critical Limitations
Serum total cortisol measured by automated immunoassay is the most widely used clinical test. Its critical limitation is that it measures both bound and free cortisol — and is therefore confounded by any condition that alters CBG or albumin levels. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) offers superior specificity by chromatographically separating cortisol from structurally similar steroids before mass-specific detection, eliminating the antibody cross-reactivity inherent to immunoassays.
Immunoassays systematically overestimate cortisol relative to LC-MS/MS. Urinary free cortisol immunoassays show a mean positive bias of approximately 121–135%, and salivary immunoassay values typically run 2–2.5× higher than LC-MS/MS equivalents. This platform discordance has direct consequences for clinical thresholds applied to screening decisions.
Cross-reactivity is the second major immunoassay limitation. Prednisolone cross-reactivity ranges from 6% on some platforms to 55% on others. 11-Deoxycortisol — elevated in 21-hydroxylase deficiency or after metyrapone challenge — produces a clinically relevant 23% positive bias in certain immunoassay formats. Free cortisol can be calculated from total cortisol and CBG using validated equations, though modifications incorporating albumin improve accuracy in critical illness states.
Reference ranges are time-of-day dependent: morning (8–9 AM) serum cortisol typically 140–690 nmol/L by immunoassay; late-night or midnight serum cortisol should be <50 nmol/L for normal HPA axis suppression.
4.2 Salivary Cortisol: Free Fraction, Pre-Analytical Pitfalls, and the Cortisone Advantage
Salivary cortisol reflects the unbound cortisol fraction that passively diffuses from plasma, making it independent of CBG fluctuations. This is particularly advantageous in populations using oral contraceptives: women on combined hormonal contraceptives show morning serum cortisol values in a reference range of 284–994 nmol/L (versus 159–569 nmol/L in non-users) due to estrogen-driven CBG induction, while salivary cortisol shows no clinically significant difference between groups.
A notable refinement in salivary measurement is that salivary cortisone — produced by 11β-HSD2 in the salivary glands converting cortisol to cortisone — may be a superior surrogate for serum free cortisol (r = 0.95 with serum free cortisol). Salivary cortisone is less susceptible to pre-analytical contamination from residual blood in oral samples and is increasingly preferred in research and diagnostic settings equipped for LC-MS/MS.
Pre-analytical requirements for salivary cortisol are strict:
- Fasting or avoiding food/drink for ≥15 minutes before collection
- Avoiding vigorous exercise, smoking, or tooth brushing ≤30 minutes prior
- Objective awakening time verification for CAR protocols
- Standardised device (Salivette or equivalent)
- Samples stable at room temperature for ≤24 hours; freeze at −20°C for longer storage
4.3 24-Hour Urinary Free Cortisol: Integrated Output With Collection Caveats
The 24-hour urinary free cortisol (UFC) measurement provides an integrated measure of free cortisol secretion across the full diurnal cycle, recommended as one of three first-line biochemical screening tests for Cushing's syndrome. The LC-MS/MS reference range is approximately 11–70 μg/day (30–190 nmol/day), with immunoassay values approximately 1.5–2× higher. Key analytical caveats include:
- Falsely low results with renal impairment (eGFR <60 mL/min), as tubular reabsorption is enhanced
- Falsely elevated values with fluid intake exceeding 5 L/day (dilution effect on cortisol-binding proteins) or overcollection
- Requirement for complete 24-hour collection verified by co-measurement of urinary creatinine
- Individual day-to-day variability of ±20–30%, necessitating repeat collections if screening is borderline
4.4 Hair Cortisol: Retrospective Exposure Over Months
Hair cortisol concentration (HCC) provides a retrospective measure of cumulative cortisol secretion, with each proximal 1 cm segment representing approximately one month of exposure. Unlike blood or salivary cortisol, HCC is unaffected by diurnal fluctuations, venipuncture-induced stress, or CBG levels, and shows high intra-individual temporal stability.
A comprehensive meta-analysis of 124 sub-samples (N = 10,289) found that chronically stressed groups showed 22% increased HCC overall and 43% elevated HCC when stress was ongoing. Anxiety disorders (particularly PTSD) showed 17% reduced HCC compared to controls. Hair bleaching reduces HCC by approximately 25%; swimming frequency and hair washing frequency show modest inverse correlations with HCC. LC-MS/MS is the most rigorous quantification method, though validated ELISA kits are widely used in research settings.
4.5 The CBG Confounder: Why Total Cortisol Can Mislead
Cortisol-binding globulin is perhaps the most underappreciated confounder of total cortisol interpretation. CBG varies significantly across physiological states:
- Increased: 2–3-fold elevation in pregnancy and with estrogen-containing drugs (oral contraceptives, HRT); also elevated in hypothyroidism, anorexia nervosa
- Decreased: Liver cirrhosis, critical illness/sepsis (via neutrophil elastase cleavage), hyperthyroidism, nephrotic syndrome, obesity
A clinically instructive example: in surgical patients, postoperative CBG decreased 30% while total cortisol increased 55% — but the free cortisol index increased 130%, revealing that total cortisol grossly underestimated the true biological signal. This discordance is particularly consequential in ICU settings where adrenal insufficiency assessment via total cortisol alone may lead to diagnostic errors. In these contexts, salivary or calculated free cortisol is the preferred metric.
5. Biomarker Mapping: An Integrated Measurement Framework
Cortisol biomarker assessment is best approached as a multi-tier framework, with primary, secondary, and contextual analytes:
Primary cortisol biomarkers:
- Serum AM/PM cortisol — Immunoassay or LC-MS/MS; circadian snapshot, total cortisol; clinical screening for hypercortisolism and hypocortisolism
- Salivary cortisol / cortisone (CAR) — Free fraction; HPA axis reactivity and circadian amplitude; stress research and ambulatory monitoring
- 24-hour urinary free cortisol — Integrated daily secretion; LC-MS/MS preferred; Cushing's syndrome screening
- Hair cortisol concentration — Cumulative retrospective exposure; LC-MS/MS or validated ELISA; chronic stress quantification
Secondary contextual biomarkers:
- DHEA-S — Adrenal androgen reserve; anti-glucocorticoid function; the cortisol:DHEA-S ratio is a superior predictor of mortality, immune vulnerability, and epigenetic ageing compared to cortisol alone
- ACTH — Upstream HPA axis regulation; essential for differential diagnosis of cortisol excess or deficiency (ACTH-dependent vs. ACTH-independent)
- Cortisone (salivary) — 11β-HSD2 activity marker; superior free cortisol surrogate in estrogen-exposed populations
- CRP / IL-6 — Inflammatory burden; contextualises glucocorticoid resistance when elevated alongside cortisol
- CBG — Corrects total cortisol interpretation in altered-binding states
6. The Cortisol:DHEA-S Ratio: A Superior Ageing and Stress Biomarker
DHEA-S exerts anti-glucocorticoid effects by inhibiting GR nuclear translocation and blocking 11β-HSD1 activity, functionally opposing many of cortisol's catabolic and immunosuppressive actions. DHEA-S peaks in the third decade of life and declines approximately 2% per year thereafter (a process termed "adrenopause"), while cortisol remains relatively stable — producing a steadily increasing cortisol:DHEA-S ratio with advancing age.
This ratio has emerged as a superior biomarker in multiple high-stakes contexts:
- In a cohort of 4,255 Vietnam-era veterans, the cortisol:DHEA-S ratio predicted all-cause mortality over 15 years while cortisol alone did not
- The ratio was significantly elevated in treatment-resistant depression compared to treatment-responsive depression
- Elevated cortisol:DHEA-S predicted post-injury infection rate in elderly hospitalised patients
- A recent analysis across multiple epigenetic ageing clocks identified the cortisol:DHEA-S ratio as the best predictor of biological age acceleration
These findings suggest that measuring adrenal function within a longevity context benefits substantially from including DHEA-S alongside cortisol, rather than cortisol in isolation.
7. Glucocorticoid Resistance: Resolving the Cortisol-Inflammation Paradox
A central puzzle in stress biology is why chronically stressed individuals simultaneously show elevated cortisol and elevated inflammatory markers — seemingly contradictory given cortisol's immunosuppressive properties. The glucocorticoid resistance (GCR) model resolves this paradox.
Chronic stress downregulates GR expression and signalling sensitivity in immune cells, so that cortisol can no longer effectively suppress NF-κB-driven inflammation. Studies have demonstrated that GCR predicts increased susceptibility to viral infection and higher local pro-inflammatory cytokine production (IL-1β, TNF-α, IL-6). At the genomic level, chronic stress produces a fingerprint of blunted GR signalling and upregulated NF-κB signalling in peripheral blood mononuclear cells.
Population-level evidence confirms that flattened diurnal cortisol slopes mediate the relationship between perceived stress and systemic inflammation — including elevated CRP, IL-6, and fibrinogen — independent of health behaviours and demographics. Approximately 25–30% of individuals with major depression show CRP values above 3 mg/L, a threshold associated with immune-driven symptom profiles and may predict differential treatment response.
8. How Biostarks Can Help
Cortisol assessment is inherently multi-dimensional. A single morning serum cortisol value — absent knowledge of collection time, CBG status, assay platform, and co-analyte context — provides limited actionable information. A meaningful evaluation of HPA axis function and chronic stress burden requires consideration of circadian dynamics, measurement matrix, and the broader endocrine and inflammatory context.
The Biostarks Metabolic Health panel incorporates biomarkers directly relevant to glucocorticoid-driven metabolic risk, including glucose regulation, lipid profiles, and inflammatory markers that contextualise cortisol's downstream metabolic effects. The Longevity NAD⁺ panel includes DHEA-S, enabling the cortisol:DHEA-S ratio to be calculated — a more powerful marker of biological ageing and immune vulnerability than cortisol alone. For individuals seeking a comprehensive nutritional and hormonal baseline, understanding the interaction between cortisol, nutrient status, and adrenal function provides a more complete picture of systemic health.
References
- Dynamics of ACTH and Cortisol Secretion and Implications for Disease — Endocrine Reviews — Lightman, Birnie & Conway-Campbell — (2020) — Source
- Immune Regulation by Glucocorticoids — Nature Reviews Immunology — Cain & Cidlowski — (2017) — Source
- Chronic Stress, Glucocorticoid Receptor Resistance, Inflammation, and Disease Risk — PNAS — Cohen et al. — (2012) — Source
- Stress, Adaptation, and Disease: Allostasis and Allostatic Load — Annals of the New York Academy of Sciences — McEwen — (1998) — Source
- Cortisol as a Biomarker of Mental Disorder Severity — Psychosomatic Medicine — Stetler & Miller — (2011) — Source
- Analysis of Cortisol in Hair — State of the Art and Future Directions — Brain, Behavior, and Immunity — Stalder & Kirschbaum — (2012) — Source
- Measuring Cortisol in Serum, Urine and Saliva — Are Our Assays Good Enough? — Annals of Clinical Biochemistry — El-Farhan, Rees & Evans — (2017) — Source
- The Diagnosis of Cushing's Syndrome: An Endocrine Society Clinical Practice Guideline — Journal of Clinical Endocrinology & Metabolism — Nieman et al. — (2008) — Source
- Cortisol Binding Globulin: More Than Just a Carrier? — PMC / Endocrinology — Hammond — (2011) — Source
- Assessment of the Cortisol Awakening Response: Expert Consensus Guidelines — Psychoneuroendocrinology — Stalder et al. — (2016, updated 2022) — Source
- Cortisol Metabolism and the Role of 11β-Hydroxysteroid Dehydrogenase — Best Practice & Research: Clinical Endocrinology & Metabolism — Tomlinson & Stewart — (2001) — Source
- Perceived Stress Is Linked to Heightened Biomarkers of Inflammation via Diurnal Cortisol — Brain, Behavior, and Immunity — Knight et al. — (2021) — Source
