Last reviewed February 27, 2013
Many of my patients (and even more of my relatives) tell me that their arthritis pain is affected by weather. More remarkably, for many the joint symptoms precede any obvious weather change — often my patients say they are better at predicting weather than the weatherman. Could that be true, or is it another well-accepted, untested medical myth? Maybe the claim is just a commentary on the accuracy of our meteorologists. But there have been attempts to study how weather affects symptoms of joint disease. The results may hold clues into how our joints work, how to treat the pain and why there is such variability in symptoms over time.
The Case For Weather Affecting Arthritis Pain
As I learned about joint disease during my medical training, no one taught me that weather mattered. Yet I heard it so often from my patients that I wondered whether there was something to it. After all, the lack of convincing evidence and the fact that it is not taught in medical schools do not prove that there is no connection between weather and arthritis: that is, "absence of proof is not proof of absence." So, besides the stories of individuals who are absolutely convinced about weather's role in arthritis, what is the evidence that weather affects joint pain or that joint symptoms can actually predict the weather?
The medical literature dating back at least to 1859 describes theories (though no proof) that changes in barometric pressure at least partly explained changes in joint pain. In addition:
- A study published in 1929, and several others since then, asked arthritis patients about their symptoms and found that during periods of increasing barometric pressure joint pain increased.
- A similar study published in 1990 found just the opposite — that joint pain increased when barometric pressure was falling. Other, more recent trials have found small effects of humidity, pressure, temperature, the amount of sunlight, or combinations of these.
- Some studies suggest that the effects of weather on arthritis are immediate; others say there is a delayed effect of up to a month or more. For example, a 1997 study published in the Journal of Rheumatology found that the ESR, a blood test that often reflects the presence of joint inflammation, closely followed the average daily temperature from 30 days earlier; thus an increase in the temperature seemed to predict a rise in the ESR (indicating an increase in arthritis activity) 30 days later. Similarly, joint pain increased with rising temperature and improved with falling temperature but the symptom changes occurred 48 days later.
The Case Against Weather Affecting Arthritis Pain
Studies finding a link between changes in weather and changes in joint symptoms have almost all relied on "subjective" measures of arthritis — that is, pain, reported by patients, increases or decreases with some a particular weather feature. There are far fewer studies relying on examinations by physicians or results of blood tests that might be more reliable measures. In addition, for each study finding an effect, there seems to be another showing the opposite effect or no effect.
Is It All Intuition?
A study in the April 1996 Proceedings of the National Academy of Science found no association between arthritis (measured by patient report of pain or by the observation of swelling by an examiner) and weather, even though the patients thought that a strong relationship existed. More remarkably, when the data was presented in graphic form to 97 college students, many thought correlations existed when none did (according to standard statistical analysis). As summed up by the authors of the study, "… the intuitive notion of association differs from the statistical concept of association." They attributed this phenomenon to a common human attribute called "selective matching," in which we tend to remember when two things occur or change together but remember less when they do not.
For example, many people assume that they are allergic to particular foods because they became ill soon after eating it, neglecting the many times before when they had eaten that food without a problem. Although some may represent true allergies, it is likely that for many the illness was just a coincidence. Similarly, people with arthritis may notice what the weather is doing when their arthritis is worse, but take less notice of the weather when their arthritis is stable or improving.
Weather and Arthritis — Does It Matter?
Some would argue that even if weather did affect arthritis, there would be little practical importance to proving it. We cannot readily change the weather, so we are left with whatever options are currently available for the treatment of arthritis. However, some people with arthritis move to a climate they think will be better for their condition or have received that advice from their doctors. More importantly, if we did figure out the exact way weather affects the joints, it could lead to better treatments to counteract that effect. For example, if a particular chemical in the joint increased whenever the barometric pressure rose and that led to an increase in joint inflammation a day or two later, perhaps a medication to counteract that chemical could be taken as soon as the weather report mentions rising pressure.
The Bottom Line
It has not yet been proven that weather reliably affects arthritis, although it has not been disproved, either. Attempts to study the effects of weather are fraught with difficulty. There are many different types of arthritis and many different features of the weather that could act in combination to have an impact on joint symptoms. In fact, it is not clear that a definitive study is even possible. Given the widespread belief that weather matters for arthritis pain, it's probably best to assume there may be an effect that we do not yet understand. If it is true that arthritis sufferers can predict the weather, it might be the only good thing about having arthritis.
Robert H. Shmerling, M.D. is associate physician at Beth Israel Deaconess Medical Center and associate professor at Harvard Medical School. He has been a practicing rheumatologist for over 20 years at Beth Israel Deaconess Medical Center. He is an active teacher in the Internal Medicine Residency Program, serving as the Robinson Firm Chief. He is also a teacher in the Rheumatology Fellowship Program.