Exposure: Time in Nature

Pooling nine cohort studies of 8,324,652 people in seven countries, each 0.1 rise in satellite-measured vegetation within 547 yards (500 meters) of home carried 4% lower all-cause mortality (HR 0.96, 95% CI 0.94 to 0.97). Heterogeneity between the studies was severe and two of the nine found no association at all, so the pooled figure is a rough central tendency, not a stable dose-response. The unit is a greenness index, not the same as living near a park, and greener neighborhoods differ from less green ones in income, air quality, noise, safety and access to care, all of which affect mortality on their own.

Across the 16 randomized trials inside a review of 50 studies, structured outdoor nature activities improved depressed mood (SMD -0.64, 95% CI -1.05 to -0.23) and positive affect (SMD 0.95, 95% CI 0.59 to 1.31), with a large but statistically fragile effect on anxiety (SMD -0.94, upper CI near zero) and a smaller effect on negative affect (SMD -0.52). Programs of 8 to 12 weeks with 20-to-90-minute sessions did best, and physical health outcomes moved much less. Risk of bias was low to moderate in the randomized subset and moderate to high in the 34 non-randomized studies. Every program bundles the outdoors with exercise, a group, a leader and a routine, and none of the trials separates the setting from the rest of the package.

943,027 Danes born 1985 to 2003, with satellite greenness measured around every childhood address up to age 10. Those in the lowest green-space band carried up to 55% higher risk of later psychiatric disorders than those in the highest, strongest for depression and stress-related disorders, after adjustment for parental income, education, employment, psychiatric history and urbanization. Cumulative green space across childhood predicted more strongly than any single year, which fits an accumulating exposure and also fits a stable family staying in one place. Intellectual disability and schizoaffective disorder showed no clear association, and the authors still name green-space quality, crime rates and neighborhood disadvantage as unmeasured.

A nationally representative survey of 19,806 adults in England. Compared with no time in nature, 120 minutes a week or more was associated with better self-reported health (OR 1.59) and wellbeing (OR 1.23), plateauing between 200 and 300 minutes, with no difference between one long visit and several short ones. It is a single cross-sectional snapshot with both exposure and outcome reported by the same person, so healthier people getting outside more explains part of it. The association held among older adults and people with long-term conditions, and survived adjustment for local green space, neighborhood, physical activity and demographics.

Of 25 included studies, most compared the same activity, usually a walk or run, in a natural setting against a built or indoor one. Pooled self-reported emotion favored the natural setting; the attention advantage did not hold after adjusting for pre-walk differences between groups, and blood pressure and cortisol showed no consistent difference. Nobody can be blinded to which environment they are in, and the outcome that moved most is the self-reported one, the most open to expectation. The review dates from 2010, and the design problem it named is still present.

Thirty-five quantitative studies of lakes, rivers, canals and coast, mostly European. The balance of evidence favored better mental health and wellbeing (12 studies) and higher physical activity (13 studies), with general health, obesity and cardiovascular outcomes less consistent. No pooling was possible because exposure was measured differently in almost every study, from straight-line distance to the coast to self-reported visits, and coastal and waterside housing tracks income the way green neighborhoods do. The reviewers call for longitudinal work before causation is read into it.

Twenty-two studies giving 76 comparisons found a positive pooled effect of gardening across a wide spread of outcomes, including lower depression, anxiety and body mass index, and higher life satisfaction and sense of community. Egger's test detected publication bias, and the effect held at a smaller size after a trim-and-fill correction. Many comparisons are before-and-after in the same people with no control group, and gardening supplies physical work, daylight and often company alongside the plants.

38 healthy urban adults, 18 of them women, mean age 27, randomly assigned to a 90-minute 3.3 mile (5.3 km) walk through oak grassland or along a four-lane road. Self-reported rumination fell in the grassland group and did not change in the road group, and blood perfusion in the subgenual prefrontal cortex fell alongside it. Heart rate and respiration did not differ between groups, so the effect was not a difference in exertion. Nineteen people per group, one walk, no follow-up, so it says nothing about a lasting change, and rumination was self-reported by people who knew which walk they had taken.

Twenty studies covering 2,257 participants, most in Asia and Europe. Shinrin-yoku reduced symptoms in the short term, with the clearest signal on anxiety and a weaker one on depression, and no study followed participants beyond a few weeks. The authors found indications of publication bias in nearly every analysis they ran, and the included studies are small, short and mostly without an active comparison, so the pooled figures likely sit above the true effect.

Records of 46 gallbladder-surgery patients at one suburban Pennsylvania hospital between 1972 and 1981, matched into 23 pairs. Patients whose window faced trees left hospital sooner, took fewer strong painkillers, and drew fewer negative comments in nurses' notes than their matched pair facing a brick wall. Room assignment was not random; which room a patient got depended on what was free when they arrived, and rooms that face grounds differ from rooms that face a wall in ward position, noise and light as well as view. It is quoted far more often than anything else in this field and has not been repeated at scale in the four decades since.

Pooling 49 outcome measures from 42 studies across eight cognitive domains, exposure to natural environments improved working memory and cognitive flexibility at low to moderate effect sizes, with attentional control less reliable and the other five domains flat. Real places produced larger effects than photographs or video, though the real-place exposures also ran longer, which the authors say they cannot separate from the setting. Most exposures last minutes and the outcomes are laboratory tasks, so this does not show a change in how someone works the next day.

Five randomized trials, four randomized by school, covering 10,733 children aged six to nine. Adding outdoor time cut two-year myopia incidence from 26.7% to 22.5% (RR 0.84, 95% CI 0.72 to 0.98) and slowed the refractive shift by 0.13 diopters over two years; the single three-year trial reported 30.5% against 39.8%. Certainty was moderate at two years and the three-year result rests on one trial, and neither children nor teachers can be blinded to extra time outside. The exposure is daylight and distance viewing, not woodland.

Ninety-two studies identified, 28 pooled. Referral programs lowered systolic blood pressure by 4.8 mmHg (95% CI -8.9 to -0.7) and diastolic by 3.8 mmHg, improved depression scores (SMD -0.50) and anxiety scores (SMD -0.57), and added about 900 steps a day (95% CI 790 to 1,010). Most included studies carry a moderate to high risk of bias, the programs range from a walking group to a therapeutic garden, and someone who accepts a referral to spend time outdoors differs from someone who declines one.

In the same review, weekly time in moderate physical activity did not improve (mean difference 25.9 minutes, 95% CI -10.3 to 62.1), while daily step counts in the same programs rose by about 900. The two disagree because steps usually come from a device and weekly moderate activity is usually recalled on a questionnaire; either the added walking sat below moderate intensity or the questionnaires did not capture it.

The measured literature is individual studies of eight to fifty-eight people, mostly pilots, reporting changes in inflammation markers, heart-rate variability and blood measures after skin contact with the ground or a grounded surface. A large share comes from the same few authors, and several of the review authors have financial ties to a company that sells earthing products and funded the research. The specific electron-transfer mechanism has not been tested at any scale that would settle it.

Eight studies pooled out of 22 reviewed. Salivary cortisol was lower in forest groups than urban groups after the visit (MD -0.05 ug/dl, 95% CI -0.06 to -0.04), and also lower before the visit began (MD -0.08 ug/dl, 95% CI -0.11 to -0.05), a larger gap. The groups already differed before anyone entered a forest, the review's own authors attribute part of the effect to anticipation, and heterogeneity after the intervention was 88%.

Twelve healthy Japanese men aged 35 to 56 took a three-day forest trip and, separately, a matched three-day city trip. Natural killer activity rose about 50% after the forest trip and stayed raised more than seven days, did not rise after the city trip, and urinary adrenaline fell after the forest trip only. Alpha-pinene and beta-pinene were measurable in forest air and near-absent in city air. Twelve people, no randomization, no blinding, fixed trip order, and almost the entire body of this work comes from one group at Nippon Medical School; a companion study in 13 nurses found the same rise in women without a city comparison. A three-day trip away from a Tokyo office differs from an ordinary day in many ways besides the trees, and natural killer activity is a laboratory measure.