The first British Society for Rheumatology pain guideline for inflammatory arthritis calls for less reliance on long-term analgesics and more evidence-based non-drug care.
Rheumatologists should move away from long-term analgesic prescribing and towards multimodal, individualised care for people with inflammatory arthritis pain, according to new British guidance.
The 2026 British Society for Rheumatology guideline recommends routinely assessing pain, treating active inflammatory disease appropriately, and making greater use of exercise, psychological interventions, and other non-pharmacological approaches.
It also takes a cautious position on opioids, gabapentinoids, and other neuromodulators, reflecting what the authors describe as a mismatch between current prescribing and the evidence base.
Guideline Working Group co-chair, Dr Nicholas Shenker, welcomed the release of the guidance.
“Pain remains an unacceptable problem for many people with inflammatory arthritis,” he said.
“This first UK guideline brings together the latest evidence to help clinicians manage that pain more effectively. We think there’s a great need for this guideline.”
UK electronic health record data show that one in four people with inflammatory arthritis were prescribed long-term opioids and one in 10 gabapentinoids in 2020, despite a lack of evidence for efficacy and known harms.
At the same time, interventions backed by trial evidence, including exercise, physical activity, and weight management, appear to be underused.
The guideline, published in Rheumatology, contains 23 recommendations covering adults, children, and young people with inflammatory arthritis. Its scope includes rheumatoid arthritis, psoriatic arthritis, axial spondyloarthritis, and juvenile idiopathic arthritis.
It was developed by a multidisciplinary group including adult and paediatric rheumatologists, GPs, pharmacists, physiotherapists, psychologists, occupational therapists, rheumatology nurses, and people with lived experience, drawing on systematic and umbrella reviews.
It recommended that pain should be assessed at every inflammatory arthritis-related appointment, or whenever patients report worsening pain, using a visual analogue scale, numerical rating scale, or verbal rating scale together with a question designed to establish whether the pain is adequately controlled.
Patients with poorly controlled pain should undergo a face-to-face assessment for inflammatory disease activity and other potential causes. Remote assessment can be offered to patients who have access to the technology, confidence in using it, and a preference for it, but poorly controlled pain identified remotely should prompt an in-person review.
The guideline stressed that pain in inflammatory arthritis was often multifactorial rather than simply a marker of active inflammation. Around 40% of people with inflammatory arthritis may have neuropathic-like pain, while an estimated 13-21% have fibromyalgia.
Poorly controlled pain should therefore trigger a holistic review that includes depression, anxiety, disturbed sleep, and functional limitations. Clinicians should also incorporate validation, empathy, and patient-centred goals into pain assessments.
“Pain in inflammatory arthritis is not solely a biomedical issue and cannot be managed effectively through medication alone,” said Professor Yeliz Prior, Guideline Working Group member, consultant occupational therapist, and professor of clinical rehabilitation.
“High-quality pain care must incorporate evidence-based, non-pharmacological interventions as a core component, not an optional extra.”
For active inflammatory arthritis, DMARDs remained central to pain management as well as disease control. Adults, children, and young people with active disease should be offered DMARDs unless contraindicated, a recommendation that received a strength-of-agreement score of 99.3%.
Short-term glucocorticoids could also be considered as bridging therapy in active rheumatoid arthritis or inflammatory arthritis affecting multiple peripheral joints until newly started or adjusted DMARDs took effect.
The guideline took a substantially more restrained approach to conventional painkillers.
Oral NSAIDs could be considered, but clinicians should aim for the lowest dose for the shortest possible duration, reassess efficacy after initiation, and consider age, comorbidities, and concomitant medications, it said.
Non-NSAID analgesics, including weak opioids, could be considered for short-term use of around one to two weeks. Patients should be counselled about potential harms, efficacy should be assessed after starting therapy, and long-term use should be avoided where possible.
The evidence underpinning opioid use was particularly limited. The guideline authors found no evidence supporting long-term opioid use or strong opioids for inflammatory arthritis pain, while the available opioid trials were short.
In pooled data, opioids produced some improvement in patient-reported global symptoms but were also associated with more adverse events. The guideline authors concluded there was limited short-term evidence for paracetamol or weak opioids and none supporting long-term use.
For patients already taking long-term analgesics, clinicians should assess both efficacy and adverse effects and support tapering where treatment was ineffective, caused unacceptable side effects, or was no longer needed. Tapering should follow shared decision-making principles.
Routine use of neuromodulators, including antidepressants, gabapentinoids, topical capsaicin, and cannabinoids, was also not recommended specifically for inflammatory arthritis pain.
Gabapentinoids could still be considered for neuropathic pain within their licensed indications, while patients with co-existing nociplastic pain such as fibromyalgia should be managed according to relevant chronic primary pain guidance.
Patients already taking long-term neuromodulators specifically for pain should similarly have treatment efficacy and adverse effects reviewed and be supported to taper when treatment is ineffective, poorly tolerated, or no longer required.
The authors identified this as a major evidence gap, calling for high-quality trials of opioids and gabapentinoids against placebo in adults with inflammatory arthritis and chronic pain. They also called for research into how best to taper long-term opioids and neuromodulators and what effect deprescribing has on pain.
In contrast, exercise and physical activity received a strong recommendation.
The evidence was reasonably consistent across inflammatory arthritis types. Of 18 systematic or umbrella reviews of exercise or physical activity in rheumatoid arthritis, 12 reported favourable effects and four reported mixed effects. Seven of eight reviews in axial spondyloarthritis reported favourable effects, while 10 of 11 reviews involving other or mixed inflammatory arthritis populations found favourable or mixed effects.
Psychological interventions such as cognitive behavioural therapy and mindfulness could also be considered, although their effects on pain intensity appeared generally modest and the evidence quality limited. Co-existing depression and anxiety should be managed according to relevant national guidance.
Other recommendations included occupational therapy for patients whose pain affects work, education or daily activities; education about the causes and management of inflammatory arthritis pain; weight-loss support for people who are overweight or living with obesity; interventions to restore healthy sleep; and fatigue management where fatigue is present.
The guideline did not recommend specific digital technologies, medical devices, or complementary therapies for inflammatory arthritis pain because of insufficient evidence.
The authors said the guideline was intended to support a multimodal approach rather than replace one drug with another.
They identified implementation as a research priority, alongside better methods for identifying the drivers of pain in individual patients and trials of weight management for people with inflammatory arthritis, chronic pain, and obesity.
The guideline is due to be considered for a full update after three years.
