In 1967, epidemiologists began following nearly eighteen thousand British civil servants working in London’s Whitehall district. These were not coal miners or chemical plant workers, or anyone whose occupation involved obvious physical risk. Most of them sat at desks and processed paperwork for Her Majesty’s Government. Yet they died at strikingly different rates depending on where they sat in the hierarchy.
The lowest-grade workers, the messengers, doorkeepers and porters, died at three times the rate of the most senior administrators, though they had the same employer and the same access to the National Health Service, in work that carried no obvious physical danger.
When researchers controlled for every risk factor that medicine in the late 1960s knew how to measure (smoking, blood pressure, cholesterol, physical activity, body weight), those factors explained only about a quarter of the gap in deaths from heart disease. The other three-quarters was blank.
Something was killing the people at the bottom of the organisational chart, and most of it wasn’t anything medicine had spent a century learning to measure. Destitution couldn’t explain it: these were salaried government employees with pensions and job security. Neither could occupational hazard or ignorance about health. And the gradient didn’t divide the poor from the rich. Each rung down the hierarchy corresponded to worse health outcomes than the rung above, in a smooth, continuous slope with no cliff between haves and have-nots.
The answer took a generation and contributions from endocrinology, immunology, neuroscience and social epidemiology to assemble, and much of it runs through a single hormone.
The alarm system
Your body has an emergency protocol. It is fast, and almost certainly older than your capacity for language. The sequence runs like this: the hypothalamus, a small region at the base of the brain, detects a threat and releases corticotropin-releasing hormone, CRH. CRH reaches the pituitary gland, which responds with adrenocorticotropic hormone, ACTH. ACTH enters the bloodstream and hits the adrenal glands, perched atop the kidneys, which flood the body with cortisol.
This is the hypothalamic-pituitary-adrenal axis, or HPA axis, and it is the second wave of the alarm. The first wave arrives within seconds: sympathetic nerves and a surge of adrenaline quicken the heart, widen the airways, sharpen alertness and push blood to the muscles. Cortisol follows over the next minutes and keeps the effort going. It mobilises glucose into the bloodstream, providing fuel for as long as the emergency lasts. Together the two systems turn down what the body doesn’t need right now: digestion, tissue repair, reproductive function and parts of the immune response. Those systems still matter, but if a predator is closing the distance between you, fighting off a virus can wait. Every suppression redirects energy from maintenance to survival, and each makes sense on one condition: that the emergency will be brief.
The body triages: adrenaline physically diverts blood flow from the digestive tract to skeletal muscle; cortisol suppresses lymphocytes and inhibits growth hormone and sex hormones.
Each of these moves is a bet that the current crisis matters more than the maintenance it interrupts, and the bet is cheap only if the crisis ends soon.
Under normal conditions, once the threat passes, cortisol circulates back to the hippocampus and the hypothalamus, where it suppresses further CRH release. The pituitary stands down, the adrenals ease back to their baseline output, and digestion, immunity and tissue repair come back online. This negative feedback loop is the system’s brake: the very hormone that launched the emergency eventually signals its own conclusion.
Cortisol also keeps a daily rhythm. It peaks shortly after waking, a burst of fuel and alertness to get you upright and functional, then declines steadily through the day to its lowest point in the evening.
Sleep is when the suppressed systems do their heaviest work: tissue repair, immune consolidation, memory processing. The rhythm builds a normal day around that schedule, with threat in the morning and recovery by nightfall.
The system is tuned for emergencies that are over in minutes. A lion charges, you run, and the episode resolves one way or another. What the HPA axis cannot do is tell a predator that will be gone in three minutes from a job that will grind on for thirty years; it answers both with the same alarm. Under a threat that never resolves, the output does not simply stay high. Over months and years it can stay raised, sag below normal or flatten into a curve that barely falls by evening. What it rarely gets is the clean stand-down the design expects.
The fire that won’t go out
In the short term, cortisol damps down inflammation and parts of the immune response. This is sensible triage: the predator comes first, and immune surveillance can catch up later.
Under prolonged stress, that bargain breaks down. Immune cells develop glucocorticoid receptor resistance: cortisol keeps arriving, but they become less responsive to it and stop hearing its anti-inflammatory signal.
With that brake gone soft, the immune system doesn’t return to its pre-stress baseline. It swings past it into chronic low-grade inflammation, with elevated interleukin-6, tumour necrosis factor-alpha and C-reactive protein, the inflammatory markers now linked to cardiovascular disease, type 2 diabetes, depression and autoimmune conditions. The inflammation is quiet, with no fever and no acute symptoms. It runs in the background for years, doing damage the person never feels until it turns into a diagnosis.
Sheldon Cohen of Carnegie Mellon University and colleagues at Britain’s Common Cold Unit in Salisbury tested the connection between psychological state and immune function directly. They gave about four hundred healthy volunteers nasal drops containing one of five live respiratory viruses (three strains of rhinovirus, respiratory syncytial virus, or a coronavirus), assessed their psychological stress by questionnaire, then quarantined them and watched who got sick. The higher a volunteer’s stress score, the more likely they were to develop a verified clinical cold, with infection confirmed by lab isolation of the virus or a rise in virus-specific antibodies. Among the least stressed volunteers, 27 per cent developed a cold; among the most stressed, 47 per cent, and the pattern was much the same whichever of the five viruses a volunteer had received.
The glucocorticoid resistance paradox
The central paradox of chronic stress immunology: cortisol is an anti-inflammatory hormone, yet prolonged stress goes along with a pro-inflammatory state. The mechanism is glucocorticoid receptor resistance. Prolonged stress leaves immune cells less sensitive to cortisol's regulatory signal, so the same hormone exerts weaker control over them. With cortisol's restraining influence diminished, pro-inflammatory cytokines such as IL-6 and TNF-α rise without adequate opposition, and the system tips into chronic activation instead of immunosuppression.The plumbing
Cortisol raises blood pressure. Briefly, that is useful, because higher pressure drives more oxygen to muscles that need to run. Held there for good, it is destructive.
Sustained cortisol elevation pushes blood pressure up and holds it there. When healthy men were given high doses of cortisol for five days, their systolic pressure rose by about twelve millimetres of mercury; in Cushing’s syndrome, where the adrenal glands overproduce cortisol for years, around four in five patients become hypertensive. Everyday chronic stress raises cortisol far less than either, but it pushes in the same direction, alongside repeated adrenaline surges that strain the vessel walls, and it keeps pushing day after day, for years.
The strain lands on the endothelium, the single-cell layer lining every blood vessel in the body. It regulates blood flow, prevents clotting, controls the exchange of molecules between blood and tissue, and dilates vessels to accommodate changes in pressure. Chronic cortisol impairs that ability to dilate, in part by inhibiting nitric oxide–mediated vasodilation. Vessels that can’t relax under load take repeated mechanical stress on their inner walls. Inflammatory cells, already primed by the glucocorticoid resistance that turns chronic stress into chronic inflammation, infiltrate the damaged wall, where lipids accumulate and plaques form. The process is called atherosclerosis, and it is the leading cause of heart attack and stroke in wealthy countries.
The chronic inflammation running in the background and the mechanical pressure bearing down on the vessels converge on the same arterial wall, each amplifying the other.
The study that couldn't explain itself
The original Whitehall study, now called Whitehall I, was led by Donald Reid and Geoffrey Rose of the London School of Hygiene & Tropical Medicine and launched in 1967 with nearly eighteen thousand male civil servants. Its central result, a death rate that tracked grade far more closely than the known risk factors could account for, arrived without a framework to interpret it. At the time, the idea that social position itself might be the mechanism, through what position does to the body rather than through poverty or behaviour, was not yet a credible hypothesis.In 1985, Michael Marmot launched Whitehall II, a new cohort of over ten thousand civil servants, now including women, and a question the original study had never thought to ask: job control. By that he meant the specific, measurable degree to which a worker determined what they did, when they did it and how, as distinct from income or occupational prestige. Marmot’s team assessed job control at two separate time points and tracked cardiovascular outcomes over the following years.
Workers whose job control stayed low had nearly double the risk of a new coronary heart disease event compared with those whose control stayed high. That held after adjustment for employment grade, smoking, cholesterol, blood pressure, body mass index and negative affectivity. With all of them accounted for, what still predicted heart disease was how much say a worker had over the job itself.
Robert Karasek had laid the theoretical groundwork in 1979 with his demand-control model of occupational stress. High demand alone, he argued, was not the danger; a surgeon faces immense pressure but wields decision-making power over every cut. The toxic combination was high demand paired with low decision latitude: responsibility without authority, the condition of someone who must respond to everything and determines nothing. The permanent secretary chose the project, set the timeline and delegated the execution. The messenger carried instructions he had no part in forming, under a deadline he couldn’t negotiate, to a destination he hadn’t selected. Both were busy; in Karasek’s terms, only one job carried the strain. The first Whitehall II analysis of job control later sharpened Karasek’s idea: in it, the heavy workload by itself did not predict heart disease at all. Lack of control did, independently of employment grade, negative mood and the classic coronary risk factors.
The organ that remembers
The same cortisol circulating through the arteries also crosses the blood-brain barrier and reaches the hippocampus, a structure densely stocked with receptors for cortisol. That sensitivity serves a purpose. The hippocampus encodes contextual memory: this place was dangerous, this situation caused harm. Doing that job fast requires sensitivity to the threat hormone, so the organ that tags experiences as dangerous evolved to respond strongly to cortisol.
That responsiveness is now its vulnerability. From the 1980s onward, Robert Sapolsky, Bruce McEwen and their colleagues documented what chronic glucocorticoid exposure does to this tissue: the branching connections of CA3 pyramidal neurons wither, the dentate gyrus produces fewer new neurons, and in severe or prolonged cases cells die outright.
The damage is not confined to lab animals. People with Cushing’s syndrome, recurrent major depression and PTSD tend to have smaller hippocampi. In the first two, the common thread is cortisol oversecretion. PTSD is the outlier, since its basal cortisol is usually normal or even low, and whether its smaller hippocampi are damage done by the disorder or a vulnerability that was there before the trauma is still an open question.
Sonia Lupien and colleagues made the timeline visible. They tracked elderly subjects over several years and found that those with chronically elevated cortisol had measurably smaller hippocampi and worse performance on hippocampus-dependent memory tasks, the explicit memory for facts and events that locates you in your own life. The further a person’s cortisol had climbed over the years, the greater the atrophy, and the memory the hippocampus serves bore the same mark.
This is how chronic stress may become self-perpetuating. Besides handling memory, the hippocampus is part of the HPA axis’s own braking system, a node in the negative feedback loop that is supposed to end the cortisol cascade. When cortisol levels rise, a healthy hippocampus detects the elevation and signals the hypothalamus to stand down and reduce CRH production. A hippocampus damaged by years of excessive cortisol sends a weaker signal. In animals, hippocampal lesions prolong the cortisol response, impairing the negative feedback that should bring the cascade back down. The cascade can persist even after the original stressor is removed, because the organ responsible for stopping it has been damaged by what it was supposed to stop. The longer the stress lasts, the harder it may become for the body to end the stress response, even if the external conditions change.
The overlap with depression is substantial. The hippocampal atrophy and HPA dysregulation documented in chronic stress conditions are also characteristic of major depressive disorder. Whether hippocampal damage causes depression, or both are downstream consequences of sustained glucocorticoid exposure, remains under active investigation. An overactive HPA axis is one of the most consistent biological findings in major depression, and depressed patients tend to have smaller hippocampi, most clearly in those with recurrent episodes. Chronic stress and major depression are different diagnoses, but at the level of the hippocampus and the HPA axis they run on much of the same machinery.
The chromosome
Every chromosome in your body ends with a telomere, a repetitive nucleotide sequence that works as a protective cap, a buffer of expendable genetic material. In most cells, each division shortens the telomere slightly. When it reaches a critical length, the cell can no longer divide safely: it either enters senescence, a kind of functional retirement, or triggers its own death. Telomerase, the enzyme that can rebuild telomere length, was discovered by Carol Greider and Elizabeth Blackburn in 1984; with Jack Szostak, whose experiments with Blackburn showed how telomeres protect chromosomes, they shared the 2009 Nobel Prize in Physiology or Medicine. In cells where telomerase is scarce, telomere length works as a rough biological countdown of the divisions left.
In 2004, Elissa Epel and Blackburn connected chronic psychological stress directly to that countdown. They recruited fifty-eight healthy premenopausal women: nineteen controls and thirty-nine who were primary caregivers for a chronically ill child, mothers whose daily lives involved sustained, unrelenting responsibility with limited control over outcomes. Among the caregivers, more years of caregiving went with shorter telomeres and lower telomerase activity in their immune cells, and with higher oxidative stress, after controlling for the mothers’ ages. In the women reporting the highest perceived stress, the protective caps on their chromosomes were shorter by roughly as much as nine to seventeen additional years of ordinary ageing would produce.
Oxidative stress, the damage done by free radicals, rose with the women’s stress and is one likely mechanism for the erosion. The women were measured once rather than followed over time, so the study shows a link more than a process. Still, the cells that looked older belonged to women who had spent years in a specific social condition: sustained caregiving without adequate control or relief.
What the gradient proves
Whitehall I left about three-quarters of the gap in heart-disease deaths unexplained. Whitehall II pointed to where much of the rest might be hiding: when Marmot’s team ranked the contributors to the grade gradient in new heart disease, low control at work came out on top — ahead of height, smoking, cholesterol and blood pressure. Behaviour still carries real weight, smoking above all: when later Whitehall II analyses tracked habits over two decades, they accounted for a large share of the gap in deaths from all causes. And the biology of chronic stress shows the routes the remainder can travel: decades of a stress response that never fully stands down, deposited as endothelial damage in the arteries, chronic inflammation in immune tissue, atrophy in the hippocampus, shortened telomeres in the chromosome. These pathways are not a finished accounting of every lost year, but they show how a rung in an office hierarchy can reach the body.
Along with money and prestige, the hierarchy distributed decision latitude: the degree to which a person determined the conditions of their own working day. Marmot’s finding from Whitehall II is that this latitude shows up in the body: how much say a worker had over the job predicted heart disease. Low decision latitude has the features that make a stressor hardest on the HPA axis. It is uncontrollable, it recurs and it seldom ends, so the negative feedback loop rarely gets to finish its work.
The demand-control model
Robert Karasek's 1979 framework classifies jobs along two dimensions: psychological demand and decision latitude. The dangerous quadrant is high demand with low control, or "job strain." Karasek's own 1981 study of Swedish working men found that low decision latitude raised the risk of heart disease, and Marmot's Whitehall II, which measured job control in more than ten thousand civil servants, sharpened the result: low control predicted coronary events independently of grade and the classic risk factors, even though, in that first five-year analysis, high demands on their own did not. With eleven years of follow-up, high demands, and above all high demands combined with low control, turned out to predict heart disease as well.Sapolsky found the same architecture in a different species entirely. Studying wild baboon troops in Kenya over decades, he documented that subordinate males showed chronically elevated basal cortisol, a stress-hormone brake that responded sluggishly to feedback, depressed levels of protective HDL cholesterol and suppressed levels of the growth factor IGF-1. In captive monkeys studied by other researchers, chronic social stress alone was enough to speed the clogging of their arteries. In the baboons, the pathology tracked social position, the predictability of social interactions and the degree of control the animal had over its daily experience, rather than how hard it worked or how much physical danger it faced; subordinate baboons do not work harder than dominant ones. In the stable hierarchies Sapolsky studied, dominant males could predict and control their social encounters, and subordinate males could not. Rank alone does not settle it: where hierarchies are unstable the top can be the more stressful place, and in another long-running study of wild baboons the alpha males carried higher stress-hormone levels than the males just below them.
The Whitehall gradient is one instance of a broader pattern. Richard Wilkinson and Kate Pickett, reviewing more than 150 studies of income inequality and health, found that income inequality, meaning the gap between top and bottom in a society regardless of its absolute wealth, predicts worse outcomes across a wide range of health measures, from life expectancy to infant mortality. The idea still has critics, but most of the clear-cut tests they reviewed came out in its favour. The relationship held even among rich countries, where more unequal nations performed worse on health indices than less unequal ones. Marmot, who chaired the WHO Commission on Social Determinants of Health, concluded in 2008 that health inequalities arise from the conditions in which people are born, grow, live, work and age, and from the systems put in place to deal with illness, and that those conditions are themselves shaped by political, social and economic forces. Diet, exercise and personal choices matter, but they are made inside those conditions, which include the texture of daily life and the degree of control a person has over the circumstances they inhabit.
Those conditions include the physical environment, the spaces in which bodies spend their days. Studying 106 residents of deprived neighbourhoods in a Scottish city, researchers measured salivary cortisol across the day and found that people living among more green space reported lower stress and tended to show a steeper, healthier decline in cortisol through the day, the downward slope that a properly functioning HPA axis produces. The link with that healthier daily slope held for men and women alike.
Green space, then, appears to belong in the cortisol equation. The nervous system reads the environment at a level below conscious decision-making: noise, crowding, the absence of horizon, the presence or absence of trees, the length and unpredictability of a commute. These structural conditions modulate the same hormonal cascade that runs from hypothalamus to adrenal gland. A person living in a treeless estate beside a motorway rarely chose it. The environment around them was planned and built with no thought for what it does to the people inside it, and their stress chemistry bears the mark: in the Scottish neighbourhoods, residents with the least green space reported the most stress and had flatter, less healthy cortisol curves.
The conditions that keep the stress response running follow the lines of power. Precarious employment, from zero-hours contracts and gig work without sick leave to the perpetual threat of termination, is the kind of unpredictable, low-control work the demand-control model flags as dangerous. Workplace hierarchies concentrate decision-making at the top and hold the majority in permanent states of demand without control. Built environments designed without green space or quiet or adequate housing can sustain a disrupted cortisol rhythm as a background condition.
Think of the messenger at the bottom of the Whitehall hierarchy, who couldn’t decide what he worked on, or when, or for what purpose. The biology predicts what his working life asked of him: the same emergency alarm a predator would trip on a savanna, raised day after day, with the cost falling on his arteries, his immune regulation, his hippocampus and his telomeres. Whitehall never measured any of those in him. It recorded only grade, risk factors and the date of death.
His organs didn’t know they were in a government office, or what grade he held. They registered only what the cascade is built to register: a threat with no end.
Gen AI Disclaimer
Some contents of this page were generated and/or edited with the help of a Generative AI.
Media
Key Sources and References
Marmot MG, Shipley MJ, Rose G. Inequalities in death — specific explanations of a general pattern? The Lancet. 1984;323(8384):1003-1006. https://doi.org/10.1016/S0140-6736(84)92337-7
Rose G, Marmot MG. Social class and coronary heart disease. British Heart Journal. 1981;45(1):13-19. https://doi.org/10.1136/hrt.45.1.13
Marmot MG, Bosma H, Hemingway H, Brunner E, Stansfeld S. Contribution of job control and other risk factors to social variations in coronary heart disease incidence. The Lancet. 1997;350(9073):235-239. https://doi.org/10.1016/S0140-6736(97)04244-X
Harvard Health Publishing. Understanding the stress response. Harvard Medical School. https://www.health.harvard.edu/healthy-aging-and-longevity/understanding-the-stress-response
Sapolsky RM, Romero LM, Munck AU. How do glucocorticoids influence stress responses? Endocrine Reviews. 2000;21(1):55-89. https://doi.org/10.1210/edrv.21.1.0389
Cohen S, Janicki-Deverts D, Doyle WJ, Miller GE, Frank E, Rabin BS, Turner RB. Chronic stress, glucocorticoid receptor resistance, inflammation, and disease risk. Proceedings of the National Academy of Sciences. 2012;109(16):5995-5999. https://doi.org/10.1073/pnas.1118355109
Cohen S, Tyrrell DAJ, Smith AP. Psychological stress and susceptibility to the common cold. New England Journal of Medicine. 1991;325(9):606-612. https://doi.org/10.1056/NEJM199108293250903
Whitworth JA, Williamson PM, Mangos G, Kelly JJ. Cardiovascular consequences of cortisol excess. Vascular Health and Risk Management. 2005;1(4):291-299. https://pmc.ncbi.nlm.nih.gov/articles/PMC1993964/
Miller GE, Chen E, Zhou ES. If it goes up, must it come down? Chronic stress and the hypothalamic-pituitary-adrenocortical axis in humans. Psychological Bulletin. 2007;133(1):25-45. https://doi.org/10.1037/0033-2909.133.1.25
Marmot MG, Smith GD, Stansfeld S, Patel C, North F, Head J, White I, Brunner E, Feeney A. Health inequalities among British civil servants: the Whitehall II study. The Lancet. 1991;337(8754):1387-1393. https://doi.org/10.1016/0140-6736(91)93068-K
Bosma H, Marmot MG, Hemingway H, Nicholson AC, Brunner E, Stansfeld SA. Low job control and risk of coronary heart disease in Whitehall II (prospective cohort) study. BMJ. 1997;314(7080):558-565. https://doi.org/10.1136/bmj.314.7080.558
Kuper H, Marmot M. Job strain, job demands, decision latitude, and risk of coronary heart disease within the Whitehall II study. Journal of Epidemiology and Community Health. 2003;57(2):147-153. https://doi.org/10.1136/jech.57.2.147
Karasek RA. Job demands, job decision latitude, and mental strain: implications for job redesign. Administrative Science Quarterly. 1979;24(2):285-308. https://doi.org/10.2307/2392498
Karasek R, Baker D, Marxer F, Ahlbom A, Theorell T. Job decision latitude, job demands, and cardiovascular disease: a prospective study of Swedish men. American Journal of Public Health. 1981;71(7):694-705. https://doi.org/10.2105/AJPH.71.7.694
Watanabe Y, Gould E, McEwen BS. Stress induces atrophy of apical dendrites of hippocampal CA3 pyramidal neurons. Brain Research. 1992;588(2):341-345. https://doi.org/10.1016/0006-8993(92)91597-8
Sapolsky RM. Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders. Archives of General Psychiatry. 2000;57(10):925-935. https://doi.org/10.1001/archpsyc.57.10.925
Meewisse ML, Reitsma JB, de Vries GJ, Gersons BP, Olff M. Cortisol and post-traumatic stress disorder in adults: systematic review and meta-analysis. British Journal of Psychiatry. 2007;191:387-392. https://doi.org/10.1192/bjp.bp.106.024877
Lupien SJ, de Leon M, de Santi S, Convit A, Tarshish C, Nair NPV, Thakur M, McEwen BS, Hauger RL, Meaney MJ. Cortisol levels during human aging predict hippocampal atrophy and memory deficits. Nature Neuroscience. 1998;1(1):69-73. https://doi.org/10.1038/271
Pariante CM, Lightman SL. The HPA axis in major depression: classical theories and new developments. Trends in Neurosciences. 2008;31(9):464-468. https://doi.org/10.1016/j.tins.2008.06.006
Schmaal L, Veltman DJ, van Erp TG, et al. Subcortical brain alterations in major depressive disorder: findings from the ENIGMA Major Depressive Disorder working group. Molecular Psychiatry. 2016;21(6):806-812. https://doi.org/10.1038/mp.2015.69
The Nobel Prize in Physiology or Medicine 2009 — press release. NobelPrize.org. https://www.nobelprize.org/prizes/medicine/2009/press-release/
Epel ES, Blackburn EH, Lin J, Dhabhar FS, Adler NE, Morrow JD, Cawthon RM. Accelerated telomere shortening in response to life stress. Proceedings of the National Academy of Sciences. 2004;101(49):17312-17315. https://doi.org/10.1073/pnas.0407162101
Stringhini S, Sabia S, Shipley M, Brunner E, Nabi H, Kivimaki M, Singh-Manoux A. Association of socioeconomic position with health behaviors and mortality. JAMA. 2010;303(12):1159-1166. https://doi.org/10.1001/jama.2010.297
Sapolsky RM, Mott GE. Social subordinance in wild baboons is associated with suppressed high density lipoprotein-cholesterol concentrations: the possible role of chronic social stress. Endocrinology. 1987;121(5):1605-1610. https://doi.org/10.1210/endo-121-5-1605
Sapolsky RM, Alberts SC, Altmann J. Hypercortisolism associated with social subordinance or social isolation among wild baboons. Archives of General Psychiatry. 1997;54(12):1137-1143. https://doi.org/10.1001/archpsyc.1997.01830240097014
Kaplan JR, Manuck SB, Clarkson TB, Lusso FM, Taub DM, Miller EW. Social stress and atherosclerosis in normocholesterolemic monkeys. Science. 1983;220(4598):733-735. https://doi.org/10.1126/science.6836311
Sapolsky RM. The influence of social hierarchy on primate health. Science. 2005;308(5722):648-652. https://doi.org/10.1126/science.1106477
Gesquiere LR, Learn NH, Simao MC, Onyango PO, Alberts SC, Altmann J. Life at the top: rank and stress in wild male baboons. Science. 2011;333(6040):357-360. https://doi.org/10.1126/science.1207120
Wilkinson RG, Pickett KE. Income inequality and population health: a review and explanation of the evidence. Social Science & Medicine. 2006;62(7):1768-1784. https://doi.org/10.1016/j.socscimed.2005.08.036
Commission on Social Determinants of Health. Closing the gap in a generation: health equity through action on the social determinants of health. World Health Organization. 2008. https://www.who.int/publications/i/item/WHO-IER-CSDH-08.1
Roe JJ, Ward Thompson C, Aspinall PA, Brewer MJ, Duff EI, Miller D, Mitchell R, Clow A. Green space and stress: evidence from cortisol measures in deprived urban communities. International Journal of Environmental Research and Public Health. 2013;10(9):4086-4103. https://doi.org/10.3390/ijerph10094086
