13,000 Britons and their pain

The largest study of weather and pain ever done was run through phones. It found a link — and was honest about how small it was.

13,207 Britons downloaded an app to log their pain against the weather. They came from all 124 postcode areas in the UK, from one coast to the other. That makes "Cloudy with a Chance of Pain" the largest study of weather and pain ever done.

The project existed because no one could settle the question any other way. Patients have said it for generations: the joints hurt when the weather changes. But a single doctor's visit can't capture enough data to test that, and a laboratory can't reproduce a real weather change. The solution was phones, in thousands of gardens and flats, for fifteen months.

65% were still here after a week

The first figure the researchers published wasn't about pain. It was about how many stayed. 10,584 of the participants had complete baseline data and at least one pain entry. 6,850 of them — 65% — were still taking part after the first week. 4,692 — 44% — kept going after the first month.

Participants were 81% women, average age 49, with a spread of diagnoses covering arthritis (40%), rheumatoid arthritis (19%), fibromyalgia or chronic widespread pain (24%), and other pain diagnoses (23%). Retention mattered enough that the researchers published an entirely separate study about it — who stayed, and who dropped out, among participants with rheumatoid arthritis.

9,695 pain days against 81,727 without

The pain analysis itself was based on 2,658 participants: 9,695 episodes of worsened pain, weighed against 81,727 control periods without. The method is called case-crossover — each person is their own control group, compared with themselves on other days. Age, sex, diagnosis and personality drop out of the equation by themselves. The only thing left to explain the difference is the weather that day.

Humidity counted for more than pressure

For every 10 percentage point rise in relative humidity, the odds of a pain day rose (OR 1.139; 95% CI 1.099–1.181). Wind speed gave a small increase (OR 1.02 per 1 m/s). Temperature made no measurable difference (OR 0.996, not significant). Converted to standard deviations: a typical swing in humidity was linked to 12% higher odds of pain, a typical swing in pressure to around 4% lower.

With temperature held constant, the odds of a pain episode swung between roughly 0.8 and 1.2, depending on the other conditions. And humidity correlated more strongly with osteoarthritis than with the other diagnoses in the study.

Modest, and the researchers say so themselves

The authors themselves write that the effect sizes are modest, and that the findings apply to British climate — they can't simply be carried over to other latitudes. That's not a weakness of the study. It's the whole point of using numbers instead of a feeling: you get to see how small a link actually is, instead of guessing.

13,207 people downloaded an app to get an answer to a question they thought they already knew. The answer was: yes, a little, on average, in British weather. That's not the answer a headline would have chosen. It's the answer the data gave.