Nature

Gardening: exercise, therapy, or both?

Both, within limits. Ordinary gardening at moderate effort costs about 3.8 METs, roughly the same as brisk walking, digging reaches 5.0 and a hand mower 6.0, and the NHS counts pushing a mower as moderate activity and digging as strength work. On the therapy side, a meta-analysis of 22 studies found a moderate effect on health and wellbeing (Hedges' d 0.42), while a randomised trial of 291 adults ended one growing season with the gardeners 5.80 minutes a day ahead of a waiting list on moderate to vigorous activity and 1.41 g a day ahead on fibre, neither gap significant on its own, and with stress and anxiety falling further in the gardening group.

Does gardening count as exercise

Activities are rated in METs, multiples of the energy you use sitting quietly. Anything from 3 to 5.9 METs counts as moderate intensity and 6 or more as vigorous. The Compendium of Physical Activities, the list researchers use to code activity diaries, gives general gardening at moderate effort 3.8 METs. Weeding and cultivating run from 3.8 at light to moderate effort through 4.5 at moderate effort to 5.0 with a hoe, raking lawn or leaves is 4.0 and watering the garden standing or walking 4.0, planting seedlings or shrubs 4.3, planting trees 4.5, and walking behind a power mower 5.0. Digging, spading and composting sit at 5.0 METs, 3.5 at light to moderate effort and 7.3 at vigorous. A hand mower is 6.0, listed as vigorous.

The NHS guideline for adults aged 19 to 64 asks for at least 150 minutes of moderate activity a week, or 75 minutes of vigorous, plus strengthening work for all major muscle groups on at least 2 days. Its examples include pushing a lawn mower under moderate activity and heavy gardening, such as digging and shovelling, under strengthening. So 30 minutes of steady garden work on 5 days meets the aerobic target, and 2 sessions of digging or barrow work cover the strength side, as the strength after 40 guide sets out.

MET values describe a task done without pause, while an afternoon in the garden mixes effort with standing and fetching, so count only the minutes you are breathing harder: you can talk but not sing. The values are averages, not your own pulse; the Heart Rate Zones tool gives your range.

What 22 studies say about gardening and health

The best overview is a meta-analysis by Soga, Gaston and Yamaura in Preventive Medicine Reports, 2017. The authors searched PubMed in January 2016 for studies comparing gardeners with non-gardeners, or health before and after a gardening programme, and found 22 studies published after 2001 with 76 separate comparisons. Nine came from the United States, 7 from Europe, 5 from Asia and 1 from the Middle East, and the samples were small, from 14 to 514 people with a mean of 117. Outcomes ranged from depression and anxiety to cortisol, body mass index and life satisfaction.

Pooled across everything, gardening had a positive effect of Hedges' d 0.42 (95 percent confidence interval 0.36 to 0.48). By the usual convention 0.2 is small, 0.5 medium and 0.8 large, so this is a moderate effect. Wellbeing outcomes such as mood and life satisfaction moved more (d 0.47 across 58 comparisons) than health measures such as body mass index (0.31 across 18). Horticultural therapy, meaning structured sessions led by a therapist over weeks to months, reached d 0.61 across 33 of the 76 comparisons, while everyday and community gardening came in at 0.31 across the other 43. Patients gained more (0.61) than people without a diagnosis (0.32), and before-and-after measurements showed larger changes (0.60) than gardener versus non-gardener comparisons.

Two limits temper the numbers. Egger's test showed publication bias, meaning small positive studies were more likely to reach print; a trim-and-fill correction shrank the effect to d 0.35, still clearly positive. And 44 of the 76 comparisons simply set gardeners against people who do not garden, a design that cannot separate the effect of gardening from the kind of person who takes it up.

What the randomised trial in Colorado found

Comparisons of gardeners and non-gardeners cannot show cause, so the Community Activation for Prevention (CAPS) trial in Colorado randomised people instead, by the authors' account the first randomised controlled trial of community gardening in a group this mixed in age, ethnicity and income. Between 2017 and 2019 the researchers screened 493 adults on garden waiting lists who had not gardened in the past 2 years and assigned 291 of them at random, 145 to a free plot of about 10 square metres with seeds, seedlings and an introductory course, and 146 to stay on the list. Mean age was 41.5 years, 82 percent were women and 34 percent Hispanic. Diet was measured with 3 unannounced dietary recalls and activity with 7 days of accelerometry, in spring, at harvest and again the following winter. The results appeared in The Lancet Planetary Health in 2023.

What reached significance was the pattern over the season, not the gap at the end. None of the 7 prespecified primary outcomes differed significantly between the groups at harvest. Fibre intake fell in both groups, from 22.02 g a day to 21.48 among the gardeners and from 22.42 to 20.07 in the control group, while moderate to vigorous activity rose from 52.83 to 54.92 minutes a day among the gardeners and fell from 54.86 to 49.12 in the control group. Those time-by-group effects were significant (p 0.034 for fibre, p 0.012 for activity), but the harvest gaps themselves, 1.41 g of fibre and 5.80 minutes a day, were not (p 0.25 and p 0.11, against 99.5 percent confidence intervals running from a 2.09 g decrease to a 4.92 g increase and from 4.44 fewer to 16.05 more minutes). Body mass index (time by group p 0.99), waist circumference (p 0.31), sedentary time (p 0.47), combined fruit and vegetable intake (0.46 servings a day apart at harvest, p 0.21) and diet quality did not move. Perceived stress and anxiety, both secondary outcomes, fell further in the gardening group between spring and harvest, by an estimated 3.14 points against 1.12 on the PSS-10 stress scale and 2.15 against 0.89 on the GAD-7 anxiety scale for people at the 75th percentile of baseline scores; those p values, 0.025 and 0.044, belong to the between-group difference in change, and by the winter visit the difference had gone (p 0.58 and p 0.49).

A process evaluation in Scientific Reports in 2024 shows what the plot involved: of the 145 people given one, 97 (67 percent) gardened the whole season, spending a median of 90 minutes a week at the garden, with a range from 0 to 840 minutes. The limits are plain: diet, stress and anxiety were self-reported, the trial ran in one metropolitan area over one season, and the pandemic cut the last wave short, so data collected after 1 February 2020 were excluded. One allergic reaction was the only adverse event recorded.

How to make garden time count, and where to take care

Half an hour of digging, raking, mowing or carrying loads on 5 days reaches the 150 minutes; the 90 minutes a week the trial gardeners managed is a realistic start. Alternate heavy and light jobs so you keep moving for 10 minutes at a stretch, and put digging or barrow work on 2 separate days for strength. Fibre is the second lever: in the trial it fell in both groups but less among the gardeners, so plant what you will eat, and the 30 g of fibre guide shows how far beans, peas and leafy greens get you.

If you have heart disease or take blood pressure medication, heavy digging in heat combines exertion, fluid loss and sun in a way your doctor should know about; start with 10 to 15 minutes and keep water within reach. Cover up, use a sunscreen of at least factor 30, and take the middle of the day, roughly 11am to 3pm in a temperate summer, in the shade. Backs come next: digging, lifting bags of compost and long spells of stooping load the lower spine, so bend at the hips and knees, hold the load close, kneel to weed, and change task every 20 minutes rather than digging for 2 hours.

Soil is not sterile. If you are pregnant, wear gloves and wash your hands after handling soil or compost, which can carry toxoplasmosis. Bagged compost and potting mixes can carry Legionella bacteria, which cause pneumonia when the dust is breathed in, so open bags at arm's length in the open air, damp the dust down and wash your hands; if fever and a cough follow within 2 weeks, tell your doctor you handled compost. Soil and manure also carry tetanus, so a dirty wound from a thorn or a tool is worth a same-day call to your practice if your vaccination is not up to date; schedules differ by country. Gardening is not a treatment for depression or anxiety on its own; in the studies above it lowered scores alongside care, not instead of it. For mood, see the guide on green space and mental health.

Making garden time count

Frequently asked

Does gardening count towards 150 minutes a week?

Yes, for the minutes you are working at moderate effort. The NHS lists pushing a lawn mower as moderate activity and digging or shovelling as strengthening, and most garden tasks rate 3.8 to 6.0 METs, the band that runs from brisk walking to vigorous effort. Standing and looking do not count, so time the effort, not the visit.

How much gardening a week is worth doing?

In the CAPS trial the new gardeners spent a median of 90 minutes a week at their plot and measured 5.80 more minutes of moderate to vigorous activity a day than the waiting list at harvest, a gap that was not statistically significant (p 0.11). Five sessions of 30 minutes of steady work would reach the 150 minutes a week the NHS asks of adults aged 19 to 64.

Does gardening help with anxiety or low mood?

The pooled effect across 22 studies was Hedges' d 0.42, a moderate benefit, and in the randomised trial stress and anxiety, both secondary outcomes, fell further in the gardening group between spring and harvest (p 0.025 and p 0.044), most of all in those who started out most stressed, though the difference had gone by the winter visit. That supports gardening as part of care, not as a replacement for treatment; if low mood lasts more than 2 weeks, see your doctor.

Will gardening help me lose weight?

Not measurably in one season. The 291-person trial found no change in BMI or waist circumference (time by group p 0.99 and p 0.31), and the gardeners' lead of 5.80 minutes a day of moderate to vigorous activity at harvest was itself not significant (p 0.11). The meta-analysis pooled body mass index with its other health measures at d 0.31 across 18 comparisons, but most of its comparisons set gardeners against non-gardeners, which cannot prove cause.

Is horticultural therapy different from ordinary gardening?

It is gardening led by a trained therapist in structured sessions over weeks or months, usually for patients. In the meta-analysis those programmes produced an effect of d 0.61, about twice the 0.31 seen for everyday and community gardening, though the studies were small, with 14 to 514 participants.

When should I be careful in the garden?

If you have heart disease or take blood pressure medication, heavy digging in heat is the combination to discuss with your doctor first. Protect your back by lifting from the hips and knees and kneeling to weed, wear gloves if you are pregnant, and open bags of compost in the open air so you do not breathe the dust, which can carry Legionella. If a dirty wound happens and your tetanus vaccination is not up to date, contact your practice the same day.

Sources

• Soga, Gaston and Yamaura 2017, Preventive Medicine Reports: Gardening is beneficial for health, a meta-analysis · 22 studies, 76 comparisons, Hedges' d 0.42 (95 percent CI 0.36 to 0.48); volume 5 dated 2017, published online November 2016, open access on PubMed Central
• Litt et al. 2023, The Lancet Planetary Health: Effects of a community gardening intervention on diet, physical activity and anthropometry (CAPS), randomised controlled trial · 291 adults in Denver and Aurora, author manuscript on PubMed Central
• Litt et al. 2024, Scientific Reports: A process evaluation of a randomised controlled trial of community gardening · 97 of 145 gardened the whole season, median 90 minutes a week at the plot, open access on PubMed Central
• NHS: Physical activity guidelines for adults aged 19 to 64 · 150 minutes moderate or 75 vigorous, strengthening on 2 days; lists pushing a lawn mower and heavy gardening as examples
• Compendium of Physical Activities 2024: lawn and garden activities · MET values by task, the reference list maintained by the Compendium research group

Read on

Tools that fit

This guide is for general information. It does not replace medical advice or an examination.