Massage for chronic pain: where it fits in a plan

Older adults, military veterans and people in rural areas. NCCIH notes that chronic pain grows more common with age, partly because conditions like osteoarthritis pile up over the years. National survey data show both general pain and severe pain are more common among veterans than nonveterans, and people in rural areas report chronic pain more often than people in cities.

Prevalence also varies by race and Hispanic origin. An analysis of 2010–2017 survey data found the highest rate among non-Hispanic American Indian and Alaska Native respondents and the lowest among non-Hispanic Asian respondents. Among people with a diagnosis, low-back pain and osteoarthritis lead, followed by rheumatoid arthritis, migraine, carpal tunnel syndrome and fibromyalgia. Many people have chronic pain with no clear diagnosis at all. Sometimes the problem sits in the nervous system itself, which clinicians call neuropathic pain.

The pain rarely comes alone. It tends to bring disturbed sleep, restricted movement, anxiety and depression with it, so anything that helps has to be judged against that whole cluster, not just the ache.

Complementary. A complementary approach is used alongside conventional medical care; an alternative approach is used instead of it. NCCIH's rule runs under all of its guidance: do not use an unproven product or practice to put off seeing a clinician about your pain. Its concern is delay, because a missed diagnosis can leave the cause of the pain untreated while you try other things.

NCCIH has also changed how it labels these methods. What it used to group as "mind and body practices" is now sorted by how the therapy is delivered: psychological approaches such as meditation and music therapy, physical approaches such as acupuncture, massage and spinal manipulation, and combined approaches such as yoga and tai chi.

Less than most people assume, and it depends on the condition. For low-back pain, the reviews summarized by NCCIH found only weak evidence of benefit, mostly short-term, and the American College of Physicians' 2017 guideline lists massage for acute and subacute low-back pain but not for chronic. Our back-pain guide walks through those reviews one by one.

For neck and shoulder pain, a 2013 review of 12 studies (757 participants) found massage beat inactive therapies but not other active treatments, and the shoulder benefit was short-term only. For osteoarthritis, a 2017 review of 7 trials (352 participants) found low-to-moderate quality evidence that massage reduced pain and improved function, with most of the research done on knee osteoarthritis.

Fibromyalgia produced one of the more specific findings. A 2014 review of 9 studies (404 participants) concluded that massage improved pain, anxiety and depression, but only if it continued for at least five weeks, and it did nothing for sleep disturbance.

Here the sources disagree. NCCIH's chronic pain review notes low-to-moderate quality evidence that massage may reduce arthritis pain, yet the guideline from the American College of Rheumatology and the Arthritis Foundation weakly recommends against massage for hip or knee osteoarthritis, citing weaknesses in the evidence. Both draw on the same shallow pool of 7 trials, most of them on knees. When careful reviewers split like that, any effect is probably small. On safety, the risk of harm from massage looks low; NCCIH points to rare reports of serious problems, such as a blood clot, a nerve injury or a bone fracture, often tied to vigorous techniques like deep tissue work or to people already at higher risk of injury.

Acupuncture, tai chi and yoga each have a stronger case than massage for chronic pain. A large meta-analysis of individual patient data found acupuncture more effective than no treatment and more effective than sham acupuncture for back and neck pain. The gap over no treatment was wider than the gap over sham, so part of the effect comes from everything around the needle as well as the needle itself. There is moderate-quality evidence that acupuncture reduces migraine frequency, and it is the one approach NCCIH flags as having evidence it can reduce a patient's need for opioids. In studies comparing acupuncture with migraine drugs, people getting acupuncture were far less likely to drop out over side effects.

Tai chi earns the strongest guideline language of anything here. For knee or hip osteoarthritis, the American College of Rheumatology and the Arthritis Foundation strongly recommend it; people who practiced tai chi saw improvements in pain, stiffness, balance and physical function. The ACP's 2017 guideline also lists it as an option for chronic low-back pain. Yoga holds up for low-back pain in both the short and intermediate term, with effects similar to other forms of exercise, and the same 2017 guideline includes it as an option for chronic, but not acute, low-back pain.

For chronic low-back pain, the 2017 ACP guideline also points to mindfulness-based stress reduction, progressive muscle relaxation and biofeedback as reasonable first-step, nondrug options. On the psychological side, studies found mindfulness-based interventions and cognitive behavioral therapy (CBT) both lowered pain intensity and improved function, with no important difference between them, and CBT remains the prevailing psychological treatment for chronic pain.

The CDC's 2022 opioid prescribing guideline ties this together. It concluded that multiple noninvasive, nondrug treatments improve chronic pain and function, with small to moderate effects in specific conditions, and are not associated with serious harms. Modest effects and low risk are the reason to use these approaches early and in combination rather than late and alone.

Glucosamine, chondroitin, fish oil and vitamin D for osteoarthritis, plus Reiki and magnets. The rheumatology guideline strongly recommends against glucosamine at the hand, hip or knee, because the best-designed studies show it works no better than placebo. It also advises against chondroitin and combination glucosamine-chondroitin products for the hip or knee, and against fish oil and vitamin D for osteoarthritis generally. Reiki has no high-quality research showing it relieves pain. Static magnets, the kind sold in bracelets and shoe insoles, have no conclusive evidence behind them for any type of pain.

Cannabis products sit in a murkier zone. Short-term studies suggest some formulations with THC and CBD may reduce chronic pain, with side effects like dizziness and drowsiness, and little is known about long-term use or other formulations. Hypnosis may help, but the studies that found a benefit involved at least 8 sessions, so plan for a course rather than one appointment.

With your clinician, as one piece of the plan, and as a course rather than a single visit. Massage for chronic pain tends to give a small, often short-term benefit alongside conventional care, and in the studies that tested how much massage people got, the benefit came from longer sessions repeated over several weeks.

NCCIH's most useful instruction is also its plainest: make these decisions with the people treating you. In practice, that comes down to four steps.