By employing participants' smartphones, researchers hone in on bad weather-pain connection

Patients with long-term pain were more likely to log high-pain events in their apps on days with certain weather conditions, such as high humidity.
By Dave Muoio
01:48 pm

It’s an age-old question: do joint and bone pains really get worse with the weather, or is it all just grumblings and happenstance? For an answer, a new study published in NPJ Digital Health looked to smartphones, tasking thousands participants with long-term pain to report their daily pain levels through an app and referencing these against their GPS-based local weather.

By doing so, researchers from the UK found a “significant yet modest” relationship between daily pain and certain weather conditions, in particular relative humidity. Further, participants’ increased reports persisted when taking their moods and physical activity into consideration.

“Despite much research examining the existence and nature of the weather–pain relationship,” the researchers wrote,” … studies have failed to reach consensus in part due to their small sample sizes or short durations (commonly fewer than 100 participants or one month or less); by considering a limited range of weather conditions; and heterogeneity in study design (e.g. the populations studied, methods for assessing pain, assumptions to determine the weather exposure and statistical analysis techniques),” the researchers wrote.

“In addition to investigating the weather–pain relationship, we successfully conducted a national smartphone study that delivered on the promise of how consumer technology can support health research.”


Among the 13,207 users who downloaded the “Cloudy with a Chance of Pain” study app, 2,658 completed the baseline and logged at least one pain event day that could be matched to a non-pain day during the same month. These participants, more than 80% of whom were female, stuck with the study for a median of 165 days. In all, the researchers logged 9,695 pain days and 81,727 matched non-pain days from these individuals over a total of 6,431 participant-months.

The odds that an individual would log a pain event rose significantly when participants experienced increases in relative humidity (odds ratio, 1.139 per 10% humidity increase; 95% confidence interval, 1.099 - 1.181) and wind speed (OR, 1.02 per 1  m s−1 increase; 95% CI, 1.005-1.035). An increase in atmospheric pressure was associated with decreased odds of a pain event (OR, 0.962 per 10-mbar pressure increase; 95% CI, 0.937-0.987). No difference was observed based on temperature changes or precipitation.

Another model expanded to include participants’ daily mood and physical activity. This found the former to very significantly impact the odds of a pain report, but still demonstrated the weather-based trends of the prior model.


The researchers recruited adult UK participants who owned a smartphone and were living with pain conditions for more than three months by advertising the study’s website, where those interested could download the study app. Participants were encouraged to use the app for six months, or longer if they were willing.

Each day, participants were asked to complete a 10-point questionnaire in the app regarding their pain, wellbeing, physical activity, mood and other related points. Researchers then compared these data to information on each individual’s local weather, which they obtained by linking hourly smartphone GPS data to nearby weather stations.


Digital trials are an increasingly promising avenue for research projects looking to cut costs and run decentralized investigations that are more convenient for participants. Additionally, “crowdsourced” designs like the one used in this weather-pain study offer researchers a substantially greater pool of potential participants to work with. This kind of approach was key to the huge numbers seen earlier this year in the Apple Heart Study, and likely to be seen again as the company cues up three more similar app-based health studies.


The researchers noted that this study’s lack of blinding and substantial drop-off from recruitment to the final cohort could suggest some biases among those who stuck it out. However, they felt that the size and design of their project still offered a leg up over similar investigations in the past.

“The study design has resolved problems of prior weather–pain studies such as small populations, short follow-up, surrogate pain outcomes, the absence of possible causal pathway variables such as mood, and assumptions about where participants were located and thus the weather to which they were exposed,” they wrote. “Overcoming these obstacles produced a large dataset that allowed us to tease out subtle relationships between weather and pain.”


The latest news in digital health delivered daily to your inbox.

Thank you for subscribing!
Error! Something went wrong!