Twenty women with fibromyalgia kept food diaries, wore trackers and gave blood. Half their calories came from ultra-processed food, but stress was what tracked their pain

Fibromyalgia is long-lasting, widespread pain that comes with poor sleep, tiredness and often low mood, and it mostly affects women. There is no blood test for it, and the nervous system of people who have it seems to turn the volume up on pain signals. People living with it hear a lot of advice about anti-inflammatory diets, sleep and exercise, but few studies have measured all of those things in the same people at the same time. A pilot study published in Nutrients on 25 October tried to do exactly that.

Researchers at Bournemouth University in England screened 55 volunteers and enrolled 20 women aged 18 to 65 with a doctor-confirmed diagnosis, between May and August 2025. Their median age was 48, and they had been diagnosed for a median of five and a half years. Each woman came in fasting for a blood test (three inflammation markers: IL-6, TNF-alpha and IL-10), filled in questionnaires on pain, stress and sleep, and took home a three-day food diary and a Fitbit to wear for two weeks. When she came back, a researcher pressed a gauge into her shoulder and shin muscles to find the point where pressure became painful, and tested how well her body could damp pain down on its own.

The diaries were scored two ways. One was a dietary inflammatory index, which weights 29 nutrients by how they have been linked to inflammation in earlier research; the group's median score was clearly on the pro-inflammatory side. The other was how much of each woman's energy came from ultra-processed food: a median of 50%. The women walked a median of about 6,100 steps a day and slept about seven and a half hours a night by the tracker, though their own sleep-quality scores were poor.

The clearest signal was not about food. Women who reported more stress had worse fibromyalgia impact scores, more signs of a sensitised nervous system and more pain right now, with correlations of 0.64 to 0.68, which is strong for this kind of study. After allowing for age and time since diagnosis, each extra point on the 40-point stress scale went with about 2.3 points more on the 100-point fibromyalgia impact questionnaire.

Diet did show up, but in the blood rather than in the pain. Women whose diets scored as more inflammatory had higher IL-6 (correlation 0.46). More ultra-processed food went with lower pain thresholds, meaning they hurt sooner under pressure, but those links (around −0.4) fell short of statistical significance. Among the blood markers, higher TNF-alpha went with lower pressure-pain thresholds (−0.55). Steps, sleep hours and heart-rate variability from the trackers showed nothing worth reporting. One result went against expectations: women who rated their sleep as worse actually showed stronger built-in pain damping.

The authors are blunt about what this can and can't say. Twenty people is very few, everything was measured at one point in time, and when the team corrected for the many comparisons they had made, none of the associations survived. IL-10 could only be measured in 13 of the 20 women, and the trackers are not lab-grade sleep or heart tests. They describe the results as estimates to plan a larger study, not evidence of cause. Still, it is a useful reminder for anyone living with fibromyalgia: diet and inflammation may be one thread, but in this group, stress was the one most tightly wound around the pain.