Fibromyalgia and exercise

What the research found.

• In the 2017 EULAR recommendations, aerobic and strengthening exercise holds the top evidence rating in the whole set of treatments. That puts it above cognitive behavioral therapy and above every drug listed. Those drugs include amitriptyline, duloxetine, pregabalin, and tramadol, and each of them is below exercise on the evidence.

  Macfarlane and colleagues, Annals of the Rheumatic Diseases, 2017

• Across 65 randomized trials and 3,764 participants, combined aerobic, strength, and stretching work reduced pain most, at a mean difference of 3.61 points on a visual scale. Strength training alone came last of the nine approaches, at 0.65. Its confidence interval crossed zero, so that approach didn't reach significance on its own.

  Yuan and colleagues, BMC Sports Science, Medicine and Rehabilitation, 2026

• The Cochrane review of aerobic exercise covered 13 trials and 839 people. Quality of life improved by 7.89 points and pain by 11.06, both on normalized 100-point scales. The fatigue estimate of 6.48 had a confidence interval ranging from 14.33 to negative 1.38, so it crossed no effect.

  Bidonde and colleagues, Cochrane Database of Systematic Reviews, 2017

• In a randomized trial of 226 adults, tai chi improved the fibromyalgia impact score by 5.5 points more than aerobic exercise. In the dose-matched comparison, where both ran twice weekly for 24 weeks, the difference was 16.2 points. That's a single-center trial with unblinded participants and a self-reported outcome.

  Wang and colleagues, BMJ, 2018

• Across 27 randomized trials and 1,785 people, aquatic exercise reduced pain with a standardized mean difference of 0.92. It improved function at 0.74 in that analysis. The certainty of that evidence was rated low, which is worth reading alongside the size of the effect.

  Martín Pérez and colleagues, Rheumato, 2026

Exercise outranks every drug in the guideline

The 2017 EULAR recommendations for managing fibromyalgia put non-pharmacological treatment first, and within that set aerobic and strengthening exercise holds the highest evidence rating of anything listed. It is above cognitive behavioral therapy, above multi-component programs, and above every drug in the document. That includes amitriptyline, duloxetine, milnacipran, tramadol, pregabalin, and cyclobenzaprine.

That ranking is worth reading for what it says. It says exercise has the best evidence behind it in a condition where nothing has spectacular evidence, rather than that exercise is a cure. Reading it as the first thing to try is right, and reading it as a promise isn't. The distance between those two readings is where a lot of disappointment comes from.

It also explains a frustration a lot of people describe, which is being told to exercise when what they wanted was a prescription. The recommendation isn't a brush-off, and it reflects the fact that the strongest trial evidence in this condition points at movement rather than at a tablet. That's an uncomfortable answer to be given and it's an honest one.

Which kind, according to the largest analysis

A network analysis of 65 randomized trials covering 3,764 participants ranked nine approaches against each other for pain. Combined aerobic, strength, and stretching work came first, at a mean difference of 3.61 points on a visual pain scale, from 2.82 to 4.40, with resistance and stretching second at 3.12 and aerobic with stretching third at 2.71. Stretching combined with balance training was the best of them for health status, at 25.13 points from 16.44 to 33.81. The differences between the top three are small.

The bottom of that ranking is as informative as the top. Aerobic exercise on its own came seventh at 1.26, stretching alone eighth at 1.12, and strength training alone came last at 0.65 with a confidence interval ranging from 1.47 to negative 0.17, which crosses no effect. Whatever is producing the benefit, it isn't a single modality performed hard.

Two cautions belong with those rankings. The authors reported publication bias on both of their main outcomes, with Egger's test at P equals 0.000 for the impact questionnaire and 0.002 for pain. The certainty of evidence was rated moderate for the impact questionnaire and low for pain, walking distance, and tender points. Rankings from an analysis like this describe a direction rather than a guarantee.

What the effect sizes look like up close

The Cochrane review of aerobic exercise pooled 13 trials and 839 people. Quality of life improved by 7.89 points on a normalized 100-point scale, pain by 11.06, physical function by 10.16, and stiffness by 7.96. Fatigue changed by 6.48 points with a confidence interval ranging from 14.33 to negative 1.38, which crosses zero, so fatigue is the outcome this evidence supports least.

Mixed exercise, meaning aerobic and resistance work together, was reviewed separately across 29 studies and 2,088 participants. Cochrane reported those as percentage improvements: 7 percent for quality of life, 5 percent for pain, 11 percent for physical function, and 13 percent for fatigue. That fatigue figure rests on a single trial of 493 people rather than on pooled evidence, which is worth knowing before leaning on it.

Aquatic exercise has its own body of work and a more recent analysis of 27 trials and 1,785 people found pain reduced at a standardized mean difference of 0.92 and function at 0.74. The certainty was rated low, and the heterogeneity reported was implausibly small for 27 unblinded trials of a physical treatment. Water isn't a lesser option, and the older Cochrane comparison of water against land found no clear advantage in either direction.

How much, and starting from where

The resistance training trials prescribed between 1 and 5 sets of 4 to 20 repetitions, twice a week, over 8 to 12 weeks, at 40 to 80 percent of a single repetition maximum. Reviewing those, the authors synthesized a recommendation of 1 to 2 sets of 4 to 20 repetitions. Mixed programs in the Cochrane review averaged 14 weeks at three sessions a week of roughly 50 to 60 minutes.

The tai chi trial gives the clearest evidence on duration. Among 226 adults, 24 weeks of tai chi beat 12 weeks by 9.6 points on the fibromyalgia impact score, while twice weekly was no better than once weekly. Longer mattered and more frequent didn't, which is an unusual finding and a useful one for anybody rationing energy.

The single most common way this goes wrong is starting at somebody else's baseline. The trials above enrolled people who were already willing to attend a supervised program, and the first week of any of them would be too much for plenty of people reading this. Beginning below what you think you can manage and building slowly isn't a lack of ambition, it's the thing that makes week twelve possible.

Safety, and what a flare afterward means

The withdrawal figures are the most useful safety data available, because a treatment people can't tolerate appears as people leaving. All-cause withdrawal was a risk ratio of 1.25 in the aerobic review and 1.02 in the mixed exercise review, so people weren't dropping out of exercise groups at higher rates than out of control groups. The resistance review reported no injuries, with the caveat that adverse effect reporting was limited.

A flare in the days after exertion is common in fibromyalgia and it isn't evidence that exercise is the wrong approach. It's evidence about the dose you just took. Reducing the amount and keeping the frequency does more good than stopping for a fortnight and starting again at the same level.

The one thing worth separating out is new pain that's confined to a single joint or a single area. Fibromyalgia is a widespread pain condition, and something local and new deserves assessment on its own terms rather than being absorbed into an existing diagnosis. That distinction protects you from having a treatable problem attributed to a condition you already have.

Common misconceptions.

Cautions specific to this condition.

• The most common reason people abandon exercise in fibromyalgia is a first week built to somebody else's baseline. The programs that held people started below what they thought they could manage.
• Expect a settling-in period, since the trials ran for 8 to 24 weeks and measured outcomes at the end of them.
• A flare after exertion is information about the amount rather than proof that exercise is wrong for you. In the trials, the programs that worked adjusted the amount rather than stopping.
• Ask about water if land-based work is intolerable, because the aquatic trials show a real effect and no clear disadvantage against land-based programs.
• Tell whoever prescribes your exercise that you have fibromyalgia and not another rheumatic disease, because the prescription that follows is genuinely different.
• Get new or changed pain assessed, particularly if it's in one joint or one area. Don't assume it belongs to the fibromyalgia.

Questions patients ask.