The hands that were tracked for years
A French cohort has measured finger joints against weather data. Osteoarthritis in the hand is one of the most sensitive places to look.
377 patients. Finger joints counted, one by one, against the weather — over several years. This is the DIGICOD cohort, recruited in France between April 2013 and June 2017, and it's one of the most detailed studies of weather and osteoarthritis there is. The average age was 66.5, 85% were women, and all had symptomatic hand osteoarthritis.
The hand is a logical place to look. Finger joints sit just under the skin, with little muscle or fat to cushion a pressure change. If the body is going to feel the weather anywhere, this is one of the best bets.
Humidity and tender joints
Humidity above 80% was linked to the number of spontaneously tender joints (p = 0.02). Compared with humidity below 68%, the odds of tender joints were 2.67 times higher at 79–85% humidity.
That's how much higher the odds of tender finger joints were at 79–85% humidity, compared with under 68%. But the finding didn't rise gradually with humidity in between — and without that gradual link, you can't rule out that it's chance.
DIGICOD cohort, 2025Pressure did the opposite
For air pressure, the study found an inverse link with the number of pressure-tender joints (p = 0.01): at pressure between 1006 and 1012 hPa, the odds of pressure-tender joints were 0.39 times the odds at other pressures. Temperature showed no significant link, and pain measured with the recognised AUSCAN tool was not significantly linked to humidity either (p = 0.14).
Why a missing dose-response matters
If a link is real and not just chance, you'd expect a dose-response: the more extreme the humidity or pressure becomes, the clearer the effect on the joints should be. A staircase effect, not a single blip.
The DIGICOD researchers didn't find that. The links turned up at specific intervals of humidity and pressure, but didn't flatten out into a pattern you could follow up or down. That's exactly the kind of finding that can arise when you test many weather variables against many symptom measures on the same dataset — some hit statistical significance by chance. Without a gradual link, you can't tell a real physiological phenomenon from noise.
The authors drew that conclusion themselves:
"our study supports the fact that weather does not appear to frankly influence symptoms in a significant way."
One study in a bigger picture
That stands in contrast to the systematic review and meta-analysis from 2023, which found air pressure and relative humidity positively correlated with pain intensity in osteoarthritis, pooled across many studies. DIGICOD is one well-run study with a large, long-running dataset — and even here, with years of follow-up on one of the body's most sensitive joints, the link disappears the moment you look for a pattern in it.
Aneroid doesn't make a diagnosis and doesn't recommend treatment. What the app can do is hold your own air pressure up against your own entries over time — and let you see whether there's a pattern for you, without having to guess.