Taking coal out of rural homes reduced respiratory symptoms

Design

Quasi-experimental difference-in-differences study exploiting staggered policy rollout, with two-way fixed effects models

Population

1,003 adults aged 40 or over in 50 rural Beijing villages reliant on coal heating; 20 villages entered the policy over four winters, 2018-2022

Primary outcome

Prevalence of respiratory symptoms, with fractional exhaled nitric oxide in a subsample

Effect

Any respiratory symptom fell 7.5 percentage points (95% CI -12.8 to -2.3); difficulty breathing -3.3 (-8.0 to 1.3), shortness of breath -3.5 (-9.3 to 2.3), cough -2.8 (-7.3 to 1.6); exhaled nitric oxide 0.3 ppb (-2.2 to 2.8)

Northern China's Clean Heating Policy restricted coal for winter heating in rural homes and subsidised cleaner alternatives. Because it rolled out village by village rather than everywhere at once, it produced something close to an experiment: villages that had not yet entered the scheme served as controls for those that had.

One thousand and three adults aged 40 or over in 50 rural Beijing villages, all reliant on coal heating at baseline, were followed across four winters from 2018 to 2022, with 20 villages entering the policy during that time. Exposure to the policy was associated with a 7.5 percentage point reduction in the prevalence of any respiratory symptom (95% CI -12.8 to -2.3). The individual symptoms moved in the same direction — difficulty breathing, shortness of breath, cough — but none reached significance alone. Fractional exhaled nitric oxide did not change at all (0.3 ppb, -2.2 to 2.8).

That last null result is informative rather than disappointing: it suggests the benefit did not run through eosinophilic airway inflammation, which is what exhaled nitric oxide measures. For India, where household biomass and coal exposure is a major driver of chronic respiratory disease in women who have never smoked, this is the rarest kind of evidence — a policy evaluated on health outcomes rather than on emissions.

• Ask about cooking and heating fuel in every patient with chronic respiratory symptoms
• Note that symptom benefit appeared without any change in exhaled nitric oxide
• The comparison is coal versus cleaner heating, not clean air versus polluted air
• Self-reported symptoms are the outcome here, with all that implies in an unblinded policy
• Record fuel exposure in the notes; it is a modifiable exposure that rarely gets written down

Why it matters

It is direct evidence that changing household fuel changes how people breathe, which is the argument clean-air policy usually has to make indirectly.

Don't overread it

Self-reported symptoms in an unblinded policy evaluation; the absence of any change in exhaled nitric oxide means no objective airway measure moved.

The statistics, in plain English

A 7.5 percentage point fall in any symptom is a composite; the components each moved less and their confidence intervals cross zero, so the composite is doing the statistical work. Symptoms were self-reported in a policy nobody was blinded to, and people who receive a subsidy for cleaner heating may report feeling better for reasons other than their airways. The difference-in-differences design handles fixed village differences but not anything that changed at the same time as the policy.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.