Stress is the part of preventive medicine that most resists clean measurement and most rewards taking seriously anyway. The concept that has held up is allostatic load — the cumulative cost of repeated adaptation, paid quietly across body systems and visible only once enough has accrued to show up as something else. The cleanest mechanistic story sits underneath it: sustained cortisol deposits visceral fat, and visceral fat carries cardiometabolic risk independent of body weight.
The pattern, not the moment
Acute stress is unremarkable. The body is built for it. What matters for long-term health is whether the system returns to baseline, and how often. A demanding day followed by genuine recovery is a different exposure to a demanding day followed by another demanding evening. The first is conditioning. The second is wear.
The cortisol–visceral fat axis
The most biologically anchored stress-to-disease pathway runs through cortisol and visceral fat — the fat stored deep around the abdominal organs. Sustained cortisol acts directly on visceral fat (locally amplified by an enzyme that regenerates active cortisol inside the fat tissue itself) and indirectly by suppressing growth-hormone and sex-hormone output. Behaviour compounds it: chronic stress drives intake of energy-dense food, which expands visceral fat, which then blunts the brain's own feedback control over the stress axis. Cumulative cortisol measured in hair is associated with waist-to-hip ratio and blood pressure, and visceral fat predicts cardiovascular disease, type-2 diabetes and dementia independently of body weight. This is the strongest single case for measuring waist circumference alongside weight.
What recovery actually looks like
Recovery is not the absence of stress. It is the presence of conditions under which the body's "rest and digest" system can do its work. Sleep is the largest part of it. Periods of low cognitive load, time outdoors, non-transactional social contact, and movement at sustainable intensities make up most of the rest. The interventions that reduce cardiovascular and metabolic stress markers in the literature are not exotic. They are the building blocks of a recoverable life.
The social variable
The mortality signal from loneliness and social isolation is large — comparable in magnitude to traditional cardiovascular risk factors — and under-acted on in clinical practice. The mechanism is straightforward: chronic, low-grade inflammation and a persistently activated stress response in the absence of social buffering. The people in a patient's life are part of the medical picture, whether or not the medical conversation acknowledges them. A single-line question about social connection belongs in any thoughtful preventive intake.
Autonomy is a clinical variable
Across occupational and population studies, the consistent finding is that workload matters less than the degree of control over the day. Two people with identical workloads but different autonomy show measurably different cardiovascular and inflammatory profiles over time. The Whitehall II civil-service studies and the Kivimäki Lancet meta-analysis converge on the same point: low job control is its own cardiovascular risk factor. Unwelcome to anyone selling a five-minute fix, but it is the finding.
What is worth tracking
Heart rate variability — the small beat-to-beat differences in resting heart rate, which reflect how recovered the nervous system is — is the most accessible quantitative window onto recovery state. A wearable-derived weekly trend, watched over months, is more useful than any single number and far more useful than comparisons between devices, which do not agree at the absolute level. Resting heart rate at sleep onset, sleep stages, and waking mood ratings round out a defensible minimum dataset. Waist circumference belongs in it too: the cheapest and most informative measure of the visceral-fat side of the cortisol picture. Single-point cortisol testing in someone who feels stressed but is otherwise asymptomatic is not informative and not recommended.
What actually moves the markers
The interventions with reproducible effect sizes are unglamorous and well-known. Regular structured exercise has an effect on mood and anxiety comparable to first-line antidepressants for non-severe depression. Mindfulness-based stress reduction and mindfulness-based cognitive therapy show moderate effects on anxiety and depression, and the latter is non-inferior to maintenance antidepressants for relapse prevention. Finnish-style sauna at 4–7 sessions a week is associated, in long-running Finnish cohorts, with substantially lower cardiovascular and all-cause mortality. Time in greenspace — a threshold of around two hours a week shows up repeatedly — moves blood pressure, cortisol and self-rated health. Slow-paced breathing at roughly six breaths per minute has small but real effects on autonomic tone.
What is not the answer
Stacking adaptogens, layering biohacks, optimising the day into a closed loop. A few well-applied basics, sustained, outperform any complicated stack. Ashwagandha has small short-term effects on subjective anxiety in industry-funded trials and a TGA hepatotoxicity advisory attached to it. The wider category of "stress supplements" has not produced a single intervention with the durability of exercise, sleep, social contact, or time outdoors. The work is to identify what is actually load-bearing in a given person's life and act on it — usually sleep, aerobic conditioning, and the social and occupational variables above.
§ For professionals — mechanisms & evidence+
Allostatic load — framework and limits
McEwen's 1998 NEJM formulation remains the most useful organising concept: glucocorticoids and catecholamines are essential for adaptation but exact a cost over time through four patterns — repeated hits, lack of habituation, prolonged response, and inadequate response. Seeman's original 10-marker index (HPA, sympathetic, cardiovascular, metabolic) has been operationalised in MacArthur, MIDUS and Whitehall II cohorts, with higher allostatic load predicting all-cause mortality, cognitive and physical decline beyond its component parts 1,2. Karlamangla 2006 showed that reduction in allostatic load tracked with lower subsequent mortality. The construct has no validated clinical cut-point; treat it as an organising concept rather than a single score.
The cortisol–visceral fat axis
The Björntorp hypothesis — psychosocial defeat → CRH-ACTH-cortisol elevation with reciprocal suppression of growth-hormone and gonadal axes → visceral adiposity, insulin resistance, hypertension — has held up across two decades of mechanistic and epidemiological work. Epel 2000 showed that women with high waist-to-hip ratio had greater cortisol reactivity to repeated lab stress. Local amplification at visceral adipose tissue occurs through 11β-HSD1, which regenerates active cortisol from cortisone and drives preadipocyte differentiation and lipoprotein lipase activation. Dallman's 2003 PNAS paper added the behavioural arm: chronic stress drives intake of palatable energy-dense food, which expands visceral fat, which blunts HPA feedback through CNS reward circuits 3.
Stalder's 2017 meta-analysis (124 samples, N≈10,289) associated hair cortisol with BMI, waist-to-hip ratio and systolic blood pressure 4; Tenk 2018 found perceived stress correlated with visceral obesity and dyslipidaemia, with visceral adiposity as the mediating node; Geiker 2018 mapped the bidirectional loop between stress, sleep disruption, hyperphagia and abdominal fat. Cushing's syndrome validates the mechanism in extreme form. Visceral fat predicts ASCVD, incident type-2 diabetes, dementia and all-cause mortality independent of BMI — the strongest case for waist circumference over BMI in preventive screening.
HPA axis patterns
Chronic stress moves through phases — early hypercortisolism, later hypocortisolism with glucocorticoid-receptor down-regulation and atrophied feedback, as seen in chronic PTSD and melancholic depression. The cortisol awakening response (50–75% rise within 30–45 min of waking) is blunted in burnout, PTSD and chronic fatigue, and exaggerated in anticipatory anxiety. Adam 2017's meta-analysis associated a flatter diurnal cortisol slope with worse outcomes across 10 of 12 health categories, most strongly for inflammation and immune outcomes 5. Burnout is recognised by the WHO as an occupational phenomenon (ICD-11) rather than a disorder; cortisol findings are heterogeneous and not diagnostic.
Autonomic nervous system and HRV
Tsuji 1994 (Framingham) showed lower HRV predicted 4-year all-cause mortality in elderly; Tsuji 1996 found a 1-SD reduction in low-frequency power associated with a 1.70× hazard for new cardiac events in disease-free subjects; Hillebrand 2013 confirmed a graded dose-response in Europace meta-analysis 6,7. Wearable PPG-based devices (Oura, Whoop, Apple Watch) are acceptable for nocturnal RMSSD trends but have poor absolute accuracy in motion or awake states; chest-strap ECG (e.g. Polar H10) is closer to research grade. Intra-individual change matters; between-device comparison does not. HRV biofeedback at ~6 breaths/min (Lehrer protocol) shows small-to-moderate RCT effects on anxiety, blood pressure and asthma symptom control.
Social determinants
Holt-Lunstad's 2010 meta-analysis (148 studies, N=308,849) associated stronger social relationships with a 50% increased likelihood of survival (OR 1.50, 95% CI 1.42–1.59) — effect comparable to smoking cessation and larger than physical activity or BMI 8. The 2015 update found mortality ORs of 1.29 (isolation), 1.26 (loneliness), 1.32 (living alone). The "comparable to 15 cigarettes/day" framing in the 2023 US Surgeon General advisory is an illustrative translation of these odds ratios rather than a precise equivalence. Mechanisms include chronic low-grade inflammation (IL-6, fibrinogen, hsCRP), sympathetic dominance, blunted vagal tone, and the Cole/Cacioppo CTRA gene-expression signature.
Job strain, control, demand
Karasek's demand-control model — high demand plus low control as "job strain" — was validated at scale in Whitehall II, where civil-service grade was inversely associated with coronary mortality independent of behavioural risk. Kivimäki 2012 (Lancet IPD meta-analysis, 13 European cohorts, N=197,473) found job strain associated with coronary heart disease at HR 1.23 (95% CI 1.10–1.37) 9. Siegrist's effort–reward imbalance model contributes a further 1.3–1.5× CHD risk, additive to demand-control. Autonomy is inversely associated with hsCRP, fibrinogen, fasting glucose and metabolic syndrome.
Interventions with durable signal
Goyal 2014's JAMA Internal Medicine meta-analysis (47 RCTs, N=3,515) found moderate evidence of improved anxiety (ES 0.38) and depression (ES 0.30) from mindfulness programmes; Kuyken 2016 IPD meta-analysis found mindfulness-based cognitive therapy non-inferior to maintenance antidepressants for recurrent depression relapse prevention 10,11. Recchia 2022's BJSM network meta-analysis found exercise comparable to antidepressants for non-severe depression, with the caveat of higher drop-out (RR 1.40) 12. Laukkanen 2015 (Kuopio cohort, N=2,315 men, 20.7-yr median follow-up) associated 4–7 Finnish-style sauna sessions/week with lower sudden cardiac death (HR 0.37), fatal CHD (HR 0.52) and all-cause mortality (HR 0.60) vs 1 session/week 13; the same group's 2018 Neurology paper found a similar dose-response for stroke. Twohig-Bennett 2018 (143 studies, ≈290 million participants) associated greenspace exposure with reduced diastolic blood pressure, salivary cortisol, and all-cause and cardiovascular mortality; White 2019 (UK MENE, N=19,806) identified a ≥120 min/week threshold for improved wellbeing 14. Sauna and greenspace evidence is observational; residual confounding (healthy-user effect) is plausible, but dose-response gradients and biological plausibility argue for partial causality.
Pharmacology — what holds up, what does not
SSRIs/SNRIs remain first-line for GAD, MDD and PTSD, not for "stress" as a category. Buspirone has a modest role in GAD without dependence liability. Benzodiazepines should be short-term only — tolerance, dependence, falls and a dementia signal. Ashwagandha trials (Lopresti 2019, Salve 2019) show small effects on anxiety and morning cortisol but are short, small, and frequently industry-funded; the TGA issued a hepatotoxicity safety advisory in February 2024. Rhodiola, L-theanine and magnesium have weak-to-modest evidence at best. CBD evidence outside specific anxiety paradigms is limited.
Australian regulatory context
MBS items 2715 and 2717 fund GP-prepared Mental Health Treatment Plans (with mental-health skills training); items 2700/2701 are the equivalent without MHST. Better Access funds 10 individual and 10 group psychology sessions per calendar year; the pandemic-era additional 10 sessions ceased on 31 December 2022. MBS item 2712 (MHTP review) was removed on 1 November 2025 — reviews are now claimed under time-based attendance items. Safe Work Australia's Model Code of Practice: Managing Psychosocial Hazards at Work (2022) is in force federally and across NSW, QLD, WA, SA and Tasmania; Victoria operates under separate OHS regulations. TGA scheduling notes: melatonin S3 for modified-release ≤2 mg in ≤30-day packs for primary insomnia in adults ≥55 (effective 1 June 2021); CBD S3 ≤150 mg/day in adults ≥18 (effective 1 February 2021), though no S3 CBD product is yet ARTG-registered. Speculative cortisol testing in asymptomatic adults is not supported by international guidance.
What is allostatic load?+
Allostatic load is the cumulative biological cost of repeated adaptation to stressors, paid quietly across cardiovascular, metabolic, immune and neuroendocrine systems and visible only when enough has accrued to show up as disease. Bruce McEwen's 1998 formulation remains the most useful framework for thinking about chronic stress, and the MacArthur, MIDUS and Whitehall II cohorts have shown that composite allostatic-load indices predict mortality and cognitive decline beyond their individual components.
How does chronic stress drive visceral fat?+
Sustained cortisol elevation acts on visceral adipose tissue directly — locally amplified by the enzyme 11β-HSD1 — and indirectly by suppressing growth-hormone and gonadal output and by promoting intake of energy-dense food. Hair-cortisol meta-analyses (Stalder 2017) associate cumulative cortisol exposure with BMI, waist-to-hip ratio and systolic blood pressure, and visceral fat in turn predicts cardiometabolic disease independently of BMI.
Is heart rate variability a useful thing to track?+
HRV gives a tractable, longitudinally trackable index of parasympathetic tone and recovery. Wearable readings are noisier than research-grade ECG but adequate for tracking individual trajectories over weeks; a sustained downward trend matters more than any single number, and between-device comparisons are unreliable. Low HRV has been associated with cardiovascular events and all-cause mortality in Framingham and ARIC.
How significant is loneliness as a health risk?+
The mortality signal from loneliness and social isolation is comparable in magnitude to traditional cardiovascular risk factors in meta-analytic work (Holt-Lunstad 2010, 2015), and the US Surgeon General's 2023 advisory associated loneliness with higher rates of coronary heart disease and stroke. The mechanism is straightforward — chronic inflammatory and sympathetic activation in the absence of social buffering.
Do meditation and mindfulness actually do anything measurable?+
Mindfulness-based stress reduction shows reproducible moderate effects on anxiety and depression in the Goyal 2014 JAMA Internal Medicine meta-analysis, and mindfulness-based cognitive therapy is non-inferior to maintenance antidepressants for recurrent depression relapse prevention (Kuyken 2016). Effect sizes are modest rather than dramatic, but the intervention is low-cost, low-risk and durable when practised consistently.
What single change reduces stress most reliably?+
Exercise. On the available evidence — including Cochrane and BJSM network meta-analyses — its effect size for mood and anxiety is comparable to first-line antidepressants for non-severe depression, with drop-out the main practical drawback. Sleep, social contact, time outdoors, and the degree of autonomy a person has over their day round out the short list of variables that consistently move the markers.
Should cortisol be measured in someone who feels stressed?+
Generally no. Single-point saliva or serum cortisol has high biological variability and low clinical actionability in asymptomatic adults; international Endocrine Society guidance restricts cortisol biochemistry to suspected Cushing's or adrenal insufficiency. Hair cortisol is a useful research tool but does not yet have a defined clinical role.
- 1. McEwen BS. Protective and damaging effects of stress mediators. New England Journal of Medicine. 1998.
- 2. Seeman TE, et al. Allostatic load as a marker of cumulative biological risk: MacArthur Studies of Successful Aging. PNAS. 2001.
- 3. Dallman MF, et al. Chronic stress and obesity: a new view of 'comfort food'. PNAS. 2003.
- 4. Stalder T, et al. Stress-related and basic determinants of hair cortisol in humans: a meta-analysis. Psychoneuroendocrinology. 2017.
- 5. Adam EK, et al. Diurnal cortisol slopes and mental and physical health outcomes: a systematic review and meta-analysis. Psychoneuroendocrinology. 2017.
- 6. Tsuji H, et al. Reduced heart rate variability and mortality risk in an elderly cohort: The Framingham Heart Study. Circulation. 1994.
- 7. Hillebrand S, et al. Heart rate variability and first cardiovascular event in populations without known cardiovascular disease: meta-analysis and dose-response meta-regression. Europace. 2013.
- 8. Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Medicine. 2010.
- 9. Kivimäki M, et al. Job strain as a risk factor for coronary heart disease: a collaborative meta-analysis of individual participant data. The Lancet. 2012.
- 10. Goyal M, et al. Meditation programs for psychological stress and well-being: a systematic review and meta-analysis. JAMA Internal Medicine. 2014.
- 11. Kuyken W, et al. Efficacy of mindfulness-based cognitive therapy in prevention of depressive relapse: an individual patient data meta-analysis. JAMA Psychiatry. 2016.
- 12. Recchia F, et al. Comparative effectiveness of exercise, antidepressants and their combination in treating non-severe depression: a systematic review and network meta-analysis of randomised controlled trials. British Journal of Sports Medicine. 2022.
- 13. Laukkanen T, et al. Association between sauna bathing and fatal cardiovascular and all-cause mortality events. JAMA Internal Medicine. 2015.
- 14. Twohig-Bennett C, Jones A. The health benefits of the great outdoors: a systematic review and meta-analysis of greenspace exposure and health outcomes. Environmental Research. 2018.