As the heat rose, Burkinabè farmers worked less hard — and 12 minutes longer per degree

Most estimates of what heat does to outdoor labour are modelled. A relationship between wet-bulb globe temperature and work capacity is derived from laboratory exposures, then applied to populations that have never been measured.

A study published in The Lancet Planetary Health on 22 December measured them. Over a full year, researchers instrumented subsistence farmers in Burkina Faso with research-grade wearables and recorded what happened to their bodies and their working days as the heat moved [s1].

The setup

The study used the Nouna Health and Demographic Surveillance System [s1]. Participants were healthy, actively practising farmers aged 20 to 45, neither overweight nor underweight, without chronic illness, not pregnant or breastfeeding, and living within 10 km of the weather station [s1].

Between 9 August 2021 and 30 August 2022, 39 households were recruited — 39 women and 39 men. One household dropped out, leaving 38 households and 76 participants, 38 men and 38 women, in the analysis [s1].

The wearables captured indoor and outdoor wet-bulb globe temperature alongside daily activity, estimated core body temperature, heart rate and GPS-tracked movement [s1]. Physical effort during fieldwork was operationalised as the physiological strain index, a 0-to-10 scale on which 10 is the highest strain, together with activity intensity, both expressed as functions of WBGT [s1].

What the bodies did

Strain went down as heat went up.

During fieldwork, the physiological strain index decreased as WBGT increased, by −0.04 per unit (95% CI −0.07 to −0.01, p=0.0056) [s1]. Men reduced their average activity intensity once WBGT exceeded 27°C [s1].

This is not a paradox; it is pacing. Workers who cannot leave the field slow down inside it, and the physiological measurement picks up the compensation rather than the exposure.

The behavioural side of the same adaptation shows up in the clock. Each 1°C rise in WBGT extended daily working time by 12.3 minutes (95% CI 2.75–21.89, p=0.013) [s1]. Participants also increased break duration and shifted work toward cooler times of day, and redistributed work across cooler months [s1].

The same amount of farming, spread thinner and later, at lower intensity.

Where the adaptation ran out

Women in the study had a higher physiological strain index and higher activity intensity during fieldwork than men — a PSI difference of 0.47 (95% CI 0.07–0.87, p=0.025) [s1].

The authors' interpretation is that women, who often balance household chores with fieldwork, have limited pacing strategies available, which heightens their vulnerability as heat stress rises [s1]. The measurement supports the pattern; the explanation is inference rather than something the wearables recorded.

That is the study's most consequential observation, and it is one that a model built on average work capacity cannot produce. Adaptation is not evenly distributed within a household. The person whose day already contains unpaid work that cannot be shifted to a cooler hour has less slack to spend on the heat.

What a study of 76 people can support

Not much on its own, and the design constrains it further. The participants were selected to be healthy, of normal weight and free of chronic illness [s1] — deliberately, to isolate the heat–effort relationship, but it means these are the workers best equipped to cope. Older farmers, those with hypertension or kidney disease, and those who are underweight are the groups where heat-related harm concentrates, and none of them are here.

One year of observation cannot separate seasonal effects from year-specific weather. And a longitudinal observational design shows what people did; it cannot show what it cost them. The paper's core caution is exactly this: these adaptations may become inadequate as climate change worsens [s1]. A strain index that falls because people are working slower says nothing about the harvest, the income or the food security that depends on the work getting done.

The comparison that shows what regulation can and cannot reach

Published in Health Affairs in December, a study of the other end of the labour spectrum measured what happens when an employer is legally obliged to act.

California adopted an outdoor heat standard in 2005 — the only US standard in place before 2022, and therefore the only one that can yet be evaluated, since five states adopted standards from 2022 onward and the data do not yet support assessing them [s2]. Researchers compared heat-related deaths among outdoor workers in California with those in neighbouring states from 1999 to 2020 [s2].

There was no decline during initial implementation, 2005–09. Estimated reductions were 33% after enforcement increased, in 2010–14, and 51% after policy revisions, in 2015–20 [s2]. The period-specific reductions were not individually statistically significant, but a Wald test indicated that the combined effect over 2010–20 was [s2].

The finding the two papers make together is uncomfortable. Heat standards appear to work — slowly, and only once enforcement follows the rule [s2]. And they operate through employers, which is a mechanism with no purchase on self-employed subsistence farmers, the group carrying the largest heat exposure in the world and answering to no one who can be inspected.

What to watch

Whether wearable-based occupational heat studies extend to workers with existing chronic conditions, and whether adaptation policy for agricultural populations in the Sahel shifts toward the household level — where, on this evidence, the pacing constraint actually sits.

This article is informational and is not medical advice.

Sources

• [s1] Eggert E, Gunga HC, Sié A, et al. "Physical effort during labour and behavioural adaptations in response to heat stress among subsistence farmers in Burkina Faso: a gender-specific longitudinal observational study." The Lancet Planetary Health, 9(12), published online 22 December 2025. https://doi.org/10.1016/j.lanplh.2025.101344
• [s2] Dean A, McCallum J. "California's Heat Standard And Heat-Related Deaths Among Outdoor Workers." Health Affairs, 44(12), December 2025. https://doi.org/10.1377/hlthaff.2025.00096