Therapies & Traditions · Bodywork practice

Massage Therapy

Massage therapy (hands-on manipulation of the body's soft tissues) · also called Massage, bodywork, Swedish massage, classical massage, deep tissue massage, sports massage, clinical massage, myofascial massage, trigger point massage, masaje, masoterapia

Massage therapy is the hands-on manipulation of the body's soft tissues, used to ease pain or to relax. In trials it has been studied for back and neck pain, anxiety, cancer-related symptoms, fibromyalgia, labour pain, preterm babies and sore muscles. Many results are short-term and the quality of evidence is often low or very low. Serious harm is rare, but blood clots, bleeding problems, cancer, pregnancy and fragile bones call for a talk with a doctor first.

General information, not medical advice. Last substantive update October 31, 2026 · Not medically reviewed. A named clinician has not signed off on this page. How this page is made

About Massage Therapy

Massage therapy is hands-on work on the body's soft tissues. In trials it gave short-term gains for some back pain, fibromyalgia, labour pain, cancer-related symptoms and sore muscles, but most evidence is low or very low quality, and for neck pain it probably made little difference against placebo. Serious harm is rare, and some conditions call for care.

What massage therapy is

NCCIH describes massage therapy as manipulating the body's soft tissues to help manage a health condition or enhance wellness. It says massage has been practised in most cultures throughout history and was one of the earliest ways people relieved pain.

NCCIH reports that 10.9 percent of US adults used massage therapy in 2022, more than double the 4.8 percent in 2002.

¹⁷

Styles and forms

NCCIH lists Swedish or classical massage as the most common form in Western countries, along with sports massage, clinical massage for specific goals such as releasing muscle spasms, and Eastern-derived traditions such as shiatsu and tui na. Those traditions have their own pages on this site.

The reviews read mostly pooled many styles together, so a result is rarely tied to one style. Three of the back pain trials used a mechanical device rather than hands, and trials in preterm infants compared moderate and light pressure.

^{17 4 9}

What trials found, by outcome

For low back pain, a Cochrane review of 25 trials found short-term gains but the authors have very little confidence in them. For neck pain, a 2024 Cochrane review of 33 trials found probably little to no difference from placebo, though trials with more and longer sessions looked better.

In cancer care, a 2025 analysis of 36 trials found less pain and anxiety and better quality of life, all very low certainty, and a 2023 analysis found less fatigue. In fibromyalgia, massage lasting 5 weeks or more eased pain, anxiety and depression but not sleep.

In labour, the Cochrane review found lower first-stage pain at low quality. In preterm infants, about 5 grams more daily weight gain. After hard exercise, less soreness for 72 hours. In pregnancy, lower anxiety and depression scores, none from low-bias trials.

^{4 5 15 12 8 13 9 6 7}

How sessions were run in trials, and choosing a therapist

The reviews rarely report session patterns in their abstracts. Where they do, the neck pain review found higher doses (8 or more sessions over four weeks, 30 minutes or longer) looked better, the fibromyalgia review found benefit in massage lasting 5 weeks or more, and the cancer fatigue review's best subgroup used 20 to 40 minute sessions twice a week for 3 to 5 weeks. These describe what was studied, not what any person should do.

NCCIH says 45 US states and the District of Columbia regulate massage and require licensing or certification. It advises asking a therapist about training, experience, credentials, the number of treatments likely needed, cost and insurance, and not delaying seeing a health care provider about a medical problem.

^{5 8 12 17}

At a glance

Traditional uses history, not proof

NCCIH says massage has been practised in most cultures throughout history and was one of the earliest methods people used to relieve pain
NCCIH describes Swedish (classical) massage as the most common form in Western countries and the core of most training programmes

Studied for research

Low back pain: short-term gains, low certainty and very little confidence from the reviewers
Neck pain: probably little to no difference against placebo, low certainty
Anxiety in people with cancer, very low certainty
Anxiety and low mood in pregnancy, very low certainty
Pain and quality of life in cancer care, very low certainty
Cancer-related fatigue, low certainty
Fibromyalgia pain, anxiety and depression, low certainty
Labour pain in the first stage, low certainty
Weight gain in preterm infants, low certainty
Delayed-onset muscle soreness after exercise, low certainty
Sleep in cancer survivors, very low certainty

Main compounds

Swedish (classical) massage
Deep tissue massage
Sports massage
Clinical massage
Massage delivered by device

Common forms

Swedish (classical) massage
Deep tissue massage
Sports massage
Clinical massage
Myofascial and trigger point work
Massage by device
Relaxation massage in pregnancy
Infant massage in neonatal units

Research links: Cochrane review 2015: massage for low-back pain (25 trials) · Cochrane review 2024: massage for neck pain (33 trials) · 2025 analysis: massage for cancer pain, quality of life and anxiety (36 trials) · Cochrane review 2018: massage in labour · 2003 review: the safety of massage therapy · NCCIH: Massage Therapy, What You Need To Know · Research Library · Genes & Biomarkers Library

What it is and its history

Accepted scientific name: Massage therapy (hands-on manipulation of the body's soft tissues)
Common names: Massage (English); Massage therapy (English); Bodywork (English); Masaje (Spanish); Masoterapia (Spanish)
Family or group: Manual (hands-on) therapy and bodywork. NCCIH describes massage therapy as manipulating the body's soft tissues to help manage a health condition or enhance wellness.
Look-alikes and adulterants: Name: Shiatsu and acupressure; Note: Pressure-point traditions with their own pages.; Name: Tui na; Note: A Chinese hands-on tradition with its own page.; Name: Chiropractic and spinal manipulation; Note: Joint manipulation, different from soft-tissue massage; the reviews link manipulation within massage to the most serious events.; Name: Reflexology; Note: Pressure on the feet or hands; has its own page.; Name: Foam rolling and massage guns; Note: Self-treatment tools; not hands-on massage by a therapist.

Traditional use and history history, not clinical proof

Tradition or era	Documented use	Source
Across cultures	NCCIH says massage has been practised in most cultures throughout history and was one of the earliest methods people used to relieve pain. It reports that 10.9 percent of US adults used massage therapy in 2022, more than double the 4.8 percent in 2002.	s-nccih-massage
Swedish or classical massage	NCCIH calls Swedish (classical) massage the most common form in Western countries and the core of most training programmes.	s-nccih-massage

Traditional use shows how it was used. It does not show that it works or is safe today.

What it includes, and how it may work

The parts and styles of the practice, and how researchers think it may act on the body. Proposed mechanisms are not proven effects in people.

Main parts and styles

Part or style	Kind	Where it is used	Activity studied	Level of evidence
Swedish (classical) massage	massage style	NCCIH calls it the most common form in Western countries. The NCCIH page read does not describe the strokes used.	Often the style used for relaxation. The reviews read did not separate it from other styles. ¹⁷	Human data
Deep tissue massage	massage style	NCCIH links rare serious effects to vigorous massage such as deep tissue massage. The page read does not define it.	The evidence read does not isolate this style. ¹⁷	Human data
Sports massage	massage style	NCCIH lists sports massage as a type. The recovery review pooled massage after hard exercise.	Used around training and competition; see the muscle soreness result. ^{17 6}	Human data
Clinical massage	massage style	NCCIH describes clinical massage as massage for specific goals such as releasing muscle spasms. Myofascial release and trigger point work are not defined on the NCCIH page read.	Targeted work for a specific problem; the reviews pooled many styles. ¹⁷	Human data
Massage delivered by device	mechanical method	The Cochrane back pain review included three trials where massage was done with a mechanical device rather than the hands.	Studied in a small number of back pain trials. ⁴	Human data

Mechanisms, receptors and pathways

Target or pathway	Gene or receptor	What was found	Level
How massage might work	Not specified in the sources read	The reviews read measured symptoms, not mechanisms. Explanations such as relaxation of muscles, calming of the stress response or the comfort of caring touch are proposals, not shown in the sources read. The muscle soreness review also measured creatine kinase, a muscle damage marker, which was lower after massage. ^{6 5}	Not established

Isolated compounds versus whole extracts

Massage is a package of touch, pressure, attention and the setting, and the trials read could not separate these. Sham massage is hard to make convincing, which is why many trials compare with usual care or no treatment.

11benefits and uses looked at

17studies and sources checked

Low certaintybest evidence rating found

7lab markers tracked in trials

7 low certainty4 very low certainty

What was looked at

The 11 benefits and uses checked for Massage Therapy, each with how strong the evidence is. Tap one to open it.

Low back painLow certainty
Neck painLow certainty
Anxiety in people with cancerVery low certainty
Anxiety and low mood in pregnancyVery low certainty
Sleep in cancer survivorsVery low certainty
Pain and quality of life in cancer careVery low certainty
Fatigue in people with cancerLow certainty
FibromyalgiaLow certainty
Pain during labourLow certainty
Weight gain in preterm infantsLow certainty
Delayed-onset muscle soreness after exerciseLow certainty

Lab markers tracked in the studies

Back pain intensityNeck pain intensityAnxiety scoresFatigue score in people with cancerPain in the first stage of labourDaily weight gain in preterm infantsMuscle soreness rating

What is Massage Therapy good for?

The main studied uses and how strong the human evidence is for each. These describe what trials found in the groups they studied. They are not advice to have massage for any condition, and each result is for the massage types and settings tested.

Low back painLow certaintyThe Cochrane review found short-term pain relief compared with inactive treatment, but not at longer follow-up. Against other treatments such as exercise it was better for pain but not function. The reviewers have very little confidence in the result. Increased pain was the most common adverse event. Neck painLow certaintyThe 2024 Cochrane review found low-certainty evidence of little to no difference against placebo in pain, disability or quality of life at up to 12 weeks. Trials with more and longer sessions looked better. Anxiety in people with cancerVery low certaintyA pooled analysis of 36 trials found lower anxiety after massage, rated very low certainty, with most studies at unclear or high risk of bias. Anxiety and low mood in pregnancyVery low certaintyPooled studies found lower anxiety and depression scores than with usual care, but none of the trials had a low risk of bias. Pain and quality of life in cancer careVery low certaintyThe same 2025 analysis found less pain and better quality of life after massage, all very low certainty. Fatigue in people with cancerLow certaintyA meta-analysis of 11 trials found a large reduction in cancer-related fatigue, strongest in breast cancer. Effects this large from small trials are often overestimates. FibromyalgiaLow certaintyMassage lasting 5 weeks or more eased pain, anxiety and depression straight afterwards in 9 small trials, but not sleep, and no follow-up data were available. Pain in the first stage of labourLow certaintyThe Cochrane review found lower pain than usual care (low quality). There was no clear difference in length of labour or caesarean births (very low quality). Weight gain in preterm infantsLow certaintyPooled trials found about 5 grams more daily weight gain, with larger gains from moderate-pressure than light-pressure massage. Muscle soreness after hard exerciseLow certaintyPooled trials found lower soreness at 24, 48 and 72 hours, higher muscle force and lower creatine kinase compared with no intervention. Sleep in cancer survivorsVery low certaintyFour small trials of massage by massage therapists found better self-reported sleep and longer sleep episodes. There is no pooled estimate.

Benefits, one by one

Each card is a research question, not a promise. Tap a goal to filter. Open a card to see the studies, doses and limits.

Low back pain

Low certainty

The Cochrane review of 25 trials found short-term pain relief from massage, but the authors have very little confidence in the result because the evidence is low to very low quality.

1 studiesThe abstract does not give session length or number. Pattern is not s…

Studies, doses and limits

What the trials found

Cochrane review (searches to 2014): 25 randomised trials, 3,096 adults with non-specific low back pain, almost all sub-acute or chronic; hands-on massage in most, a mechanical device in three.

Results

In sub-acute and chronic pain, massage beat inactive treatment (sham, waiting list or no treatment) for pain and function in the short term but not the long term. Against active treatments such as exercise or physical therapy it was better for pain in the short and long term, with no difference in function. One trial in acute pain found better pain relief than inactive controls.

Effect sizes as reported

Pain vs inactive control SMD -0.75 (95% CI -0.90 to -0.60); function SMD -0.72 (-1.05 to -0.39); pain vs active control SMD -0.37 short term and -0.40 long term.

Why this rating

Low to very low by the reviewers' own rating: trials could not be blinded and results were imprecise. The authors wrote that they have very little confidence in massage as a treatment.

Form and dose studied

The abstract does not give session length or number. Pattern is not summarised here.

Who was studied

Adults with non-specific low back pain, mostly sub-acute or chronic.

Limits

People knew they were getting massage, which can inflate self-rated pain scores. Benefit faded at longer follow-up.

Why researchers looked

Low back pain is very common and massage is one of the most popular hands-on treatments for it.

Proposed mechanism Hypothesis

Proposed: relaxing tight muscles and calming the nervous system's response to pain. Not tested in the review.

A proposed mechanism does not show that the effect happens in people.

No serious adverse events in these trials. Increased pain was the most common adverse event, reported by 1.5 to 25 percent of participants.

⁴

Neck pain

Low certainty

The 2024 Cochrane review found that, against a placebo, massage probably makes little to no difference to neck pain at up to 12 weeks, though trials with more and longer sessions looked better.

1 studiesThe review notes that most placebo trials used single sessions, which…

Studies, doses and limits

What the trials found

Cochrane review (searches to October 2023): 33 trials, 1,994 adults with sub-acute to chronic neck pain, compared with placebo, no treatment, or massage added to the same other treatment.

Results

Low-certainty evidence: little to no difference in pain, disability or quality of life against placebo, and perhaps a slight gain in reported treatment success. A subgroup analysis found a clinically important difference for higher doses (8 or more sessions over four weeks, 30 minutes or longer) as reported in the review.

Effect sizes as reported

Pain 3.43 points better than placebo on a 0 to 100 scale (95% CI 8.16 better to 1.29 worse; 403 people); Neck Disability Index 9.69 points better (95% CI 17.57 to 1.81 better; 68 people).

Why this rating

Low certainty by the reviewers' own rating: most placebo trials used only a single session, some placebos may not have been believable, and bias was common.

Form and dose studied

The review notes that most placebo trials used single sessions, which it considered suboptimal; the high-dose subgroup looked better.

Who was studied

Adults aged 18 to 70, 70 percent female; 85 percent non-specific neck pain.

Limits

Bias risk was common; placebo massage is hard to make convincing; adverse events were poorly reported.

Why researchers looked

Neck and shoulder tension is common with desk work and stress, and massage is widely used for it.

Proposed mechanism Hypothesis

Proposed: loosening tight muscles and improving local blood flow. Not tested in the review.

A proposed mechanism does not show that the effect happens in people.

Serious adverse events data were not available. Total adverse events (such as soreness, sweating or low blood pressure) were very low certainty.

⁵

Anxiety in people with cancer

Very low certainty

A 2025 pooled analysis of 36 trials found lower anxiety after massage in people with cancer, but the evidence was rated very low certainty.

1 studiesTypes of massage and session patterns are not detailed in the abstrac…

Studies, doses and limits

What the trials found

Systematic review and meta-analysis (searches to January 2024): 36 randomised trials, 3,671 people with cancer.

Results

Massage lowered anxiety and pain and improved quality of life after the intervention, but all outcomes were rated very low certainty and most studies had unclear or high risk of bias.

Effect sizes as reported

Anxiety SMD -0.38 (95% CI -0.57 to -0.18).

Why this rating

Very low by the review's own GRADE rating.

Form and dose studied

Types of massage and session patterns are not detailed in the abstract.

Who was studied

People with cancer, mixed types.

Limits

Mostly unclear or high risk of bias; effects measured straight after treatment.

Why researchers looked

Anxiety is common during cancer care and massage is one of the most requested supportive therapies.

Proposed mechanism Hypothesis

Proposed: relaxation and the comfort of caring touch. Not tested in the review.

A proposed mechanism does not show that the effect happens in people.

The abstract does not report safety. Cancer-specific cautions are on the Safety tab.

¹⁵

Anxiety and low mood in pregnancy

Very low certainty

Pooled studies in pregnant women found lower anxiety and depression scores with massage than with usual care, but none of the trials was at low risk of bias.

1 studiesNot summarised in the abstract.

Studies, doses and limits

What the trials found

Systematic review (8 studies, 7 randomised trials) of massage in pregnant women over 18.

Results

Four studies pooled for depression and five for anxiety both favoured massage over usual care.

Effect sizes as reported

Depression (CES-D scale) mean difference -5.95 (95% CI -8.11 to -3.80); anxiety SMD -0.59 (95% CI -1.06 to -0.12; I2 75 percent).

Why this rating

Compiler's rating: no trial had low risk of bias, anxiety results varied a lot, and the studies were small.

Form and dose studied

Not summarised in the abstract.

Who was studied

Pregnant women over 18.

Limits

Biased trials; self-rated outcomes; short term.

Why researchers looked

Pregnancy often brings anxiety and low mood, and drug-free options are of interest.

Proposed mechanism Hypothesis

Proposed: relaxation and supportive touch. Not tested in the review.

A proposed mechanism does not show that the effect happens in people.

The authors say more evidence on safety is needed before practice recommendations can be made.

⁷

Sleep in cancer survivors

Very low certainty

A small review of four trials found better sleep after massage given by massage therapists, but there is no pooled estimate and the evidence is thin.

1 studiesMassage was delivered by massage therapists; sessions are not detaile…

Studies, doses and limits

What the trials found

Systematic review (PubMed, Scopus and others): 4 randomised trials of massage and 3 of relaxation therapy in cancer survivors.

Results

The massage trials found better self-reported sleep and longer sleep episodes measured by an accelerometer. The relaxation trials showed no significant improvement.

Effect sizes as reported

No pooled effect size given in the abstract.

Why this rating

Compiler's rating: only four trials, no pooling, and no certainty rating in the abstract.

Form and dose studied

Massage was delivered by massage therapists; sessions are not detailed in the abstract.

Who was studied

Cancer survivors with sleep disturbance.

Limits

Four small trials; the review did not test whether non-certified providers give the same result.

Why researchers looked

Sleep problems are common after cancer treatment, and relaxation touch is one drug-free option.

Proposed mechanism Hypothesis

Proposed: relaxation. Not tested in the review.

A proposed mechanism does not show that the effect happens in people.

Not reported in the abstract.

¹¹

Pain and quality of life in cancer care

Very low certainty

A 2025 pooled analysis of 36 trials found less pain and better quality of life after massage in people with cancer, but with very low certainty.

1 studiesTypes and patterns are not detailed in the abstract.

Studies, doses and limits

What the trials found

Systematic review (searches to January 2024): 36 randomised trials, 3,671 people with cancer.

Results

Massage improved pain and quality of life after the intervention; all outcomes were rated very low certainty and most studies had unclear or high risk of bias.

Effect sizes as reported

Pain SMD -0.51 (95% CI -0.68 to -0.33); quality of life SMD 0.48 (0.19 to 0.78).

Why this rating

Very low by the review's own GRADE rating.

Form and dose studied

Types and patterns are not detailed in the abstract.

Who was studied

People with cancer, mixed types.

Limits

Bias risk; effects measured straight after treatment.

Why researchers looked

Pain and reduced quality of life are common in cancer and its treatment.

Proposed mechanism Hypothesis

Proposed: relaxation and gentle touch easing the pain experience. Not tested in the review.

A proposed mechanism does not show that the effect happens in people.

Cancer-specific cautions are on the Safety tab.

¹⁵

Fibromyalgia

Low certainty

A 2014 analysis of 9 trials found that massage lasting 5 weeks or more eased pain, anxiety and depression straight afterwards, but not sleep, with no follow-up data.

1 studiesOnly massage lasting at least 5 weeks showed benefit in this subgroup.

Studies, doses and limits

What the trials found

Systematic review (searches to June 2013): 9 randomised trials, 404 people with fibromyalgia.

Results

Massage lasting 5 weeks or more improved pain, anxiety and depression immediately afterwards but not sleep disturbance. No follow-up data were available.

Effect sizes as reported

Pain SMD 0.62 (95% CI 0.05 to 1.20); anxiety SMD 0.44 (0.09 to 0.78); depression SMD 0.49 (0.15 to 0.84); sleep SMD 0.19 (-0.38 to 0.75).

Why this rating

Compiler's rating: nine small trials, wide intervals and no follow-up.

Form and dose studied

Only massage lasting at least 5 weeks showed benefit in this subgroup.

Who was studied

Adults with fibromyalgia.

Limits

Few trials per subgroup; no long-term data.

Why researchers looked

Fibromyalgia causes widespread pain and has few effective treatments.

Proposed mechanism Hypothesis

Proposed: relaxation and reduced muscle tension. Not tested.

A proposed mechanism does not show that the effect happens in people.

Not reported in the abstract.

⁸

Pain during labour

Low certainty

The 2018 Cochrane review found low-quality evidence that massage eased pain during the first stage of labour compared with usual care, with no clear effect on length of labour or caesarean births.

1 studiesNot summarised in the abstract.

Studies, doses and limits

What the trials found

Cochrane review (searches to June 2017): 14 trials, 10 of them (1,055 women) pooled.

Results

Massage lowered first-stage pain intensity (low quality). Very low-quality evidence showed no clear difference in length of labour, pain medicine, assisted vaginal birth or caesarean section. Single trials reported less anxiety and more sense of control, and two reported more satisfaction.

Effect sizes as reported

First-stage pain SMD -0.81 (95% CI -1.06 to -0.56; 6 trials, 362 women); length of labour MD 20.64 minutes (-58.24 to 99.52); caesarean RR 0.75 (0.51 to 1.09).

Why this rating

Low to very low by the reviewers' GRADE.

Form and dose studied

Not summarised in the abstract.

Who was studied

Women in labour.

Limits

Most trials at high risk of performance and detection bias; few reported key outcomes.

Why researchers looked

Many women want drug-free ways to manage labour pain.

Proposed mechanism Hypothesis

Proposed: comfort, relaxation and sense of support. Not tested.

A proposed mechanism does not show that the effect happens in people.

Few trials reported on safety.

¹³

Weight gain in preterm infants

Low certainty

A 2020 pooled analysis of 15 trials found that preterm infants given massage gained about 5 grams more per day, with larger gains from moderate-pressure than light-pressure massage.

1 studiesA moderate-pressure versus light-pressure subgroup comparison is repo…

Studies, doses and limits

What the trials found

Meta-analysis (searches to July 2018): 15 randomised trials, 697 preterm infants of very low or low birth weight.

Results

Massage improved daily weight gain.

Effect sizes as reported

Daily weight gain 5.07 g/day (95% CI 2.19 to 7.94); moderate pressure 5.60 g/day; light pressure 1.08 g/day.

Why this rating

Compiler's rating: the abstract gives no certainty rating, and weight is a stand-in for health.

Form and dose studied

A moderate-pressure versus light-pressure subgroup comparison is reported; other details are not in the abstract.

Who was studied

Preterm infants of very low or low birth weight.

Limits

Outcome is weight; the abstract does not describe risk of bias.

Why researchers looked

Weight gain is the main thing deciding when a preterm baby can go home.

Proposed mechanism Hypothesis

The review does not say how massage might work; this is not tested here.

A proposed mechanism does not show that the effect happens in people.

Not reported in the abstract.

⁹

Delayed-onset muscle soreness after exercise

Low certainty

A 2017 pooled analysis of 11 trials found lower muscle soreness at 24, 48 and 72 hours after hard exercise in people who received massage.

1 studiesMassage types and timings differ by trial and are not detailed in the…

Studies, doses and limits

What the trials found

Systematic review (searches to December 2016): 11 randomised trials with 504 participants.

Results

Soreness ratings were lower than with no intervention at 24, 48 and 72 hours; muscle force and peak torque were higher and creatine kinase (a muscle damage blood marker) was lower.

Effect sizes as reported

Soreness SMD -1.16 overall (95% CI -1.60 to -0.72); force SMD 0.56; peak torque SMD 0.38; creatine kinase SMD -0.64.

Why this rating

Compiler's rating: self-rated soreness, small studies, different exercises and massage types pooled.

Form and dose studied

Massage types and timings differ by trial and are not detailed in the abstract.

Who was studied

Healthy people after strenuous exercise.

Limits

No certainty rating in the abstract; compared with no intervention, not a sham.

Why researchers looked

Soreness after unaccustomed or hard exercise is common, and athletes use massage for recovery.

Proposed mechanism Hypothesis

Proposed: reduced swelling or muscle tension. Not tested here.

A proposed mechanism does not show that the effect happens in people.

Not reported in the abstract.

⁶

Connections to the rest of health

How Massage Therapy connects to the rest of our health knowledge graph. Every link says what kind of connection it is. A safety link (for example to the liver) is a warning, not a benefit. A study that merely mentions a topic never creates a benefit link.

Explore every connection: goals, body systems, organs, symptoms, genes

Kind of connection Connected to Search

A. Wellness goals

Connected to	Evidence status	Context	Studies
Pain Relief	Low certainty	Low back pain: The Cochrane review of 25 trials found short-term pain relief from massage, but the authors have very little confidence in the result because the evidence is low to very low quality. Neck pain: The 2024 Cochrane review found that, against a placebo, massage probably makes little to no difference to neck pain at up to 12 weeks, though trials with more and longer sessions looked better.	^{4 5 15 8 13}
Anxiety Support	Very low certainty	Anxiety in people with cancer: A 2025 pooled analysis of 36 trials found lower anxiety after massage in people with cancer, but the evidence was rated very low certainty. Anxiety and low mood in pregnancy: Pooled studies in pregnant women found lower anxiety and depression scores with massage than with usual care, but none of the trials was at low risk of bias.	^{15 7 8}
Calm Relaxation	Very low certainty	Anxiety in people with cancer: A 2025 pooled analysis of 36 trials found lower anxiety after massage in people with cancer, but the evidence was rated very low certainty.	¹⁵
Mood Balance	Very low certainty	Anxiety and low mood in pregnancy: Pooled studies in pregnant women found lower anxiety and depression scores with massage than with usual care, but none of the trials was at low risk of bias.	⁷
Better Sleep	Very low certainty	Sleep in cancer survivors: A small review of four trials found better sleep after massage given by massage therapists, but there is no pooled estimate and the evidence is thin.	¹¹
Reduce Fatigue	Low certainty	Fatigue in people with cancer: A 2023 pooled analysis of 11 trials found a large reduction in cancer-related fatigue with massage, with the strongest effect in breast cancer.	¹²
Womens Wellness	Low certainty	Pain during labour: The 2018 Cochrane review found low-quality evidence that massage eased pain during the first stage of labour compared with usual care, with no clear effect on length of labour or caesarean births.	¹³
Muscle Recovery	Low certainty	Delayed-onset muscle soreness after exercise: A 2017 pooled analysis of 11 trials found lower muscle soreness at 24, 48 and 72 hours after hard exercise in people who received massage.	⁶

B. Body systems

Connected to	Kind of connection	Evidence status	Context	Studies
Musculoskeletal System	Biological association	Low certainty	Studied for: Low back pain Studied for: Neck pain	^{4 5}
Pain Inflammation	Biological association	Low certainty	Studied for: Low back pain Studied for: Pain and quality of life in cancer care	^{4 15 8 13}
Nervous System	Biological association	Very low certainty	Studied for: Anxiety in people with cancer Studied for: Anxiety and low mood in pregnancy	^{15 7 11 8}
Stress Mood	Biological association	Very low certainty	Studied for: Anxiety in people with cancer	¹⁵
Reproductive System	Biological association	Very low certainty	Studied for: Anxiety and low mood in pregnancy Studied for: Pain during labour	^{7 13}
Sleep Recovery	Biological association	Very low certainty	Studied for: Sleep in cancer survivors	¹¹
Whole Body	Biological association	Very low certainty	Studied for: Pain and quality of life in cancer care Studied for: Fatigue in people with cancer	^{15 12 9}
Bones Joints Muscles	Biological association	Low certainty	Studied for: Delayed-onset muscle soreness after exercise	⁶

E. Health conditions

Connected to	Kind of connection	Evidence status	Context	Studies
Blood clots (deep vein thrombosis)	Adverse effect or risk	Case report	Pulmonary embolism and blood-clot related events are listed among rare reported harms. Preparation or compound: Deep or vigorous massage	^{3 17}
Osteoporosis and fragile spine	Adverse effect or risk	Case report	Bone fracture is among rare reported harms; one case report in a man with undiagnosed ankylosing spondylitis. Preparation or compound: Vigorous or traditional massage	^{1 17}
Low back pain		Low certainty	Short-term pain relief; the reviewers have very little confidence. Preparation or compound: Massage (hand or device)	⁴
Fibromyalgia		Low certainty	Lower pain, anxiety and depression straight afterwards; no follow-up data. Preparation or compound: Massage lasting 5 weeks or more	⁸
Anxiety		Very low	Lower anxiety scores in cancer and pregnancy trials; certainty very low. Preparation or compound: Massage	^{15 7}

I. Biomarkers

Connected to	Kind of connection	Evidence status	Context	Studies
Back pain intensity	Biological association	Low certainty	Measured in trials: lower than inactive treatment in the short term; weaker over longer periods. A lab change is not the same as a health benefit.	⁴
Neck pain intensity	Biological association	Low certainty	Measured in trials: little to no difference against placebo at up to 12 weeks. A lab change is not the same as a health benefit.	⁵
Anxiety scores	Biological association	Very low certainty	Measured in trials: lower than comparison groups straight after massage. A lab change is not the same as a health benefit.	^{15 7}
Fatigue score in people with cancer	Biological association	Low certainty	Measured in trials: lower than controls. A lab change is not the same as a health benefit.	¹²
Pain in the first stage of labour	Biological association	Low certainty	Measured in trials: lower than usual care. A lab change is not the same as a health benefit.	¹³
Daily weight gain in preterm infants	Biological association	Low certainty	Measured in trials: higher with massage. A lab change is not the same as a health benefit.	⁹
Muscle soreness rating	Biological association	Low certainty	Measured in trials: lower than no intervention at 24, 48 and 72 hours. A lab change is not the same as a health benefit.	⁶

L. Active compounds

Connected to	Kind of connection	Evidence status	Context	Studies
Swedish (classical) massage	Biological association	Human	Often the style used for relaxation. The reviews read did not separate it from other styles.	¹⁷
Deep tissue massage	Biological association	Human	The evidence read does not isolate this style.	¹⁷
Sports massage	Biological association	Human	Used around training and competition; see the muscle soreness result.	^{17 6}
Clinical massage	Biological association	Human	Targeted work for a specific problem; the reviews pooled many styles.	¹⁷
Massage delivered by device	Biological association	Human	Studied in a small number of back pain trials.	⁴

Each connection is stored with the ingredient, the connected record, the kind of connection, the evidence status, the supporting studies, the preparation or compound involved, and a review date (last reviewed October 31, 2026).

Evidence at a glance: what the research found for 11 outcomes

7Low certainty

4Very low certainty

Each outcome below is one thing researchers measured. Certainty is how far the evidence can be trusted, not how well it works. The full studies are on the next tab.

Low back pain 1 study Low certainty Neck pain 1 study Low certainty Anxiety in people with cancer 1 study Very low certainty Anxiety and low mood in pregnancy 1 study Very low certainty Sleep in cancer survivors 1 study Very low certainty Pain and quality of life in cancer care 1 study Very low certainty Fatigue in people with cancer 1 study Low certainty Fibromyalgia 1 study Low certainty Pain during labour 1 study Low certainty Weight gain in preterm infants 1 study Low certainty Delayed-onset muscle soreness after exercise 1 study Low certainty

The evidence

The research centre for Massage Therapy. Every record here was checked against PubMed or the publisher's page; we never list a study we have not verified.

Evidence by outcome

Bars show how many studies we found for each outcome, not how well it works. People are added up across trials only (reviews are not added, to avoid double counting).

How we rate evidence

High certainty

Several well-run trials or reviews agree; further research is unlikely to change the picture much.

Moderate certainty

Good trials point the same way, but some doubt remains.

Low certainty

Small, short or inconsistent trials. The finding could change with better research.

Very low certainty

Very little or very weak human data. Treat as a possibility only.

Mixed findings

Trials or reviews disagree with each other.

Not enough evidence

Not enough human research to say either way.

Evidence against

Good research found no benefit, or found harm.

We take the reviewers' own risk-of-bias and GRADE statements where they exist, and otherwise judge from study size, design, consistency, directness and whether a product matches what is sold. Many publications do not make evidence high. Method note: We searched PubMed, re-read the abstract of every review and trial cited, and read the NCCIH page on massage therapy through a tool that summarises pages. Certainty ratings are the compiler's own reading, except where a review gives its own GRADE (stated in the text). Funding is not stated in most abstracts, and this is noted where relevant.

Study library: every study we checked 17

Design Research type Year from Search

Study	Design	People	Product, dose, length	Result
Abilash 2017 Fracture-dislocation at C6-C7 level with Quadriplegia after Traditional Massage in a Patient with Ankylosing Spondylitis: A Case Report Malays Orthop J · PMID 29021885 · DOI	Case report	1 One middle-aged man with undiagnosed ankylosing spondylitis (a stiffening spine disease)	Traditional massage (an Asian hands-on tradition, not Swedish massage) Not given in the abstract Not given in the abstract	A neck fracture-dislocation with complete loss of nerve function and neurogenic shock after traditional massage in a person whose spine disease had not been diagnosed. The authors say traditional massage can be dangerous without adequate knowledge and skill. One person (Not given for a case report) Methods, limits and funding Outcomes: Injury Limitations: A single case of a traditional technique in a person with a fragile spine; says nothing about how often this happens. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Corbin 2005 Safety and efficacy of massage therapy for patients with cancer Cancer Control · PMID 16062163 · DOI	Narrative review	Literature on massage in people with cancer (MEDLINE and CINAHL)	Massage therapy Not given in the abstract	The author concludes that conventional cancer care can safely incorporate massage, although people with cancer may be at higher risk of rare adverse events; strongest benefit evidence is for stress and anxiety reduction. Risks can be minimised when clinician and patient discuss it. Not given (Not given for a case report) Methods, limits and funding Outcomes: Safety and efficacy in cancer care Limitations: Narrative review from 2005; the abstract does not list the specific risks; older than the pooled analyses on the Benefits tab. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Ernst 2003 The safety of massage therapy Rheumatology (Oxford) · PMID 12777645 · DOI	Systematic review of case reports	16 case reports and 4 case series of adverse effects of massage (oil and ice effects excluded)	Any type of massage therapy Not given in the abstract Reports describe events after the fact; no set duration · vs No comparison group	Reported events included stroke (cerebrovascular accidents), haematoma, nerve damage, pulmonary embolism, pseudoaneurysm, leg ulcers and others. Most were linked to exotic types of manual massage or massage by non-therapists; serious events were mostly linked to techniques other than Swedish massage. The author concludes serious adverse events are probably rare. 20 reports (Not given for a case report) Methods, limits and funding Outcomes: Adverse effects of any type Limitations: Case reports only, from before 2003; cannot give a rate; under-reporting is likely. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Furlan 2015 Massage for low-back pain Cochrane Database Syst Rev · PMID 26329399 · DOI	Systematic review meta analysis	3096 25 randomised trials in adults with non-specific low-back pain (one acute, the rest sub-acute or chronic); in 3 trials massage was done with a mechanical device, in the rest by hand	Massage (soft-tissue manipulation by hand or mechanical device) Not given in the abstract Short-term and long-term follow-up reported; session patterns not summarised in the abstract · vs Inactive controls (sham, waiting list, no treatment) and active controls (manipulation, mobilisation, TENS, acupuncture, traction, relaxation, physical therapy, exercise, self-care education)	Quality of evidence rated low to very low. In sub-acute and chronic pain, massage beat inactive controls for pain and function in the short term but not the long term. Against active controls it was better for pain in the short and long term but not for function. No serious adverse events; increased pain intensity was the most common adverse event (1.5 to 25 percent of participants). The authors have very little confidence that massage is effective. Sub-acute/chronic pain vs inactive control: SMD -0.75 (7 studies, 761 people); function SMD -0.72 (6 studies, 725 people). Pain vs active control: SMD -0.37 short term (12 studies, 964 people); SMD -0.40 long term (5 studies, 757 people). Acute pain vs inactive control: SMD -1.24 (1 study, 51 people). (Pain vs inactive -0.90 to -0.60; function -1.05 to -0.39; pain vs active short term -0.62 to -0.13, long term -0.80 to -0.01; acute pain -1.85 to -0.64) Methods, limits and funding Outcomes: Pain, function, adverse events Limitations: Hard to blind people or therapists, so performance and measurement bias were common; downgraded for risk of bias and imprecision. Only one trial in acute pain. Funding and conflicts: The abstract says the majority of trials were funded by not-for-profit organisations. No conflict statement in the record read. Read: abstract
Gross 2024 Massage for neck pain Cochrane Database Syst Rev · PMID 38415786 · DOI	Systematic review meta analysis	1994 33 randomised trials in adults aged 18 to 70 (70 percent female) with sub-acute to chronic neck pain; 85 percent non-specific neck pain, 6 percent radiculopathy, 9 percent cervicogenic headache	Any type of massage Not given in the abstract Mostly single sessions in the placebo comparison; a subgroup analysis looked at 8 or more sessions over 4 weeks of 30 minutes or longer · vs Placebo or sham (10 trials), no treatment (8 trials), or massage added to the same co-treatment (15 trials)	Against placebo, low-certainty evidence suggests massage probably makes little to no difference to pain, disability or quality of life at up to 12 weeks, and may slightly improve reported treatment success. A subgroup analysis found a clinically important difference for higher doses (8 or more sessions over four weeks, 30 minutes or longer). Total adverse events were very low certainty (soreness, sweating, low blood pressure); serious adverse events and patient satisfaction data were not available. Pain on a 0 to 100 scale: 3.43 points better than placebo (8 studies, 403 people). Neck Disability Index: 9.69 points better (2 studies, 68 people). Adverse events RR 0.99 (2 studies, 175 people). (Pain 8.16 better to 1.29 worse; disability 17.57 to 1.81 better; adverse events RR 0.08 to 11.55) Methods, limits and funding Outcomes: Pain, function-disability, quality of life, treatment success, adverse events Limitations: Many trials had unclear allocation concealment and a placebo that may not be credible; most placebo-comparison trials used suboptimal doses. Adverse events were reported too poorly to analyse properly. Funding and conflicts: The abstract says 15 percent of trials received research-institute funding. No conflict statement in the record read. Read: abstract
Guo 2017 Massage Alleviates Delayed Onset Muscle Soreness after Strenuous Exercise: A Systematic Review and Meta-Analysis Front Physiol · PMID 29021762 · DOI	Systematic review meta analysis	504 11 articles (23 data points) of randomised trials in people after strenuous exercise	Massage after exercise Not given in the abstract Soreness measured at 24, 48 and 72 hours · vs No intervention	Soreness ratings were lower with massage than with no intervention at 24, 48 and 72 hours; muscle force and torque were higher, and creatine kinase was lower. Soreness overall SMD -1.16 (24 h -0.61; 48 h -1.51; 72 h -1.46); maximal isometric force SMD 0.56; peak torque SMD 0.38; creatine kinase SMD -0.64 (Soreness overall -1.60 to -0.72; force 0.21 to 0.90; torque 0.04 to 0.71; creatine kinase -1.04 to -0.25) Methods, limits and funding Outcomes: Muscle soreness, muscle force and torque, serum creatine kinase Limitations: The abstract gives no certainty rating or risk-of-bias summary; a wide range of massage types and exercises were pooled. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Hall 2020 The effectiveness of massage for reducing pregnant women's anxiety and depression; systematic review and meta-analysis Midwifery · PMID 32827841 · DOI	Systematic review meta analysis	8 studies (7 randomised trials) in pregnant women over 18	Massage interventions Not given in the abstract Not summarised in the abstract · vs Usual care	Pooled analysis of four studies found lower depression scores and of five studies lower anxiety scores than usual care. None of the trials had a low risk of bias. Depression (CES-D) mean difference -5.95; anxiety SMD -0.59 (Depression -8.11 to -3.80; anxiety -1.06 to -0.12 (I2 75 percent for anxiety)) Methods, limits and funding Outcomes: Self-rated anxiety and depression Limitations: No trial had low risk of bias; anxiety results varied a lot between trials; sample size is not given in the abstract. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Li 2014 Massage therapy for fibromyalgia: a systematic review and meta-analysis of randomized controlled trials PLoS One · PMID 24586677 · DOI	Systematic review meta analysis	404 9 randomised trials in people with fibromyalgia	Massage therapy lasting 5 weeks or more (subgroup) Not given in the abstract 5 weeks or more · vs Controls (not detailed in the abstract)	Massage lasting 5 weeks or more improved pain, anxiety and depression straight afterwards but not sleep disturbance; no follow-up data were available. Pain SMD 0.62 (printed as positive in the abstract); anxiety SMD 0.44; depression SMD 0.49; sleep SMD 0.19 (not significant) (Pain 0.05 to 1.20; anxiety 0.09 to 0.78; depression 0.15 to 0.84; sleep -0.38 to 0.75) Methods, limits and funding Outcomes: Pain, anxiety, depression, sleep disturbance Limitations: Few trials in each subgroup; no evidence on follow-up effects; the pain interval is wide and close to zero. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Lu 2020 Massage therapy for weight gain in preterm neonates: A systematic review and meta-analysis of randomized controlled trials Complement Ther Clin Pract · PMID 32379694 · DOI	Systematic review meta analysis	697 15 randomised trials in preterm infants of very low or low birth weight	Massage therapy (moderate-pressure and light-pressure subgroups) Not given in the abstract Not summarised in the abstract	Massage improved daily weight gain; moderate-pressure massage gave a larger gain than light-pressure massage. Daily weight gain 5.07 g/day (moderate pressure 5.60 g/day; light pressure 1.08 g/day) (Overall 2.19 to 7.94; moderate pressure 2.64 to 8.56; light pressure 0.29 to 1.86) Methods, limits and funding Outcomes: Daily weight gain Limitations: The abstract gives no risk-of-bias or certainty rating; outcome is weight, not long-term health. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Mueller 2021 Effects, Side Effects and Contraindications of Relaxation Massage during Pregnancy: A Systematic Review of Randomized Controlled Trials J Clin Med · PMID 34441781 · DOI	Systematic review	12 randomised trials of relaxation massage in pregnant women, all at least 12 weeks pregnant at the start; 7 trials excluded women with difficult pregnancies or existing complications	Relaxation massage Not given in the abstract Not summarised in the abstract · vs Standard care, or standard care plus another intervention	Benefits reported included less stress, back and leg pain, depression and anxiety. Only 2 trials reported side effects, which were minor and transient. Complications that occurred were judged most likely unrelated to massage. The review discusses precautions such as preventing pulmonary embolism. 12 trials (Not given for a case report) Methods, limits and funding Outcomes: Effects, side effects and contraindications Limitations: Authors rate methodological quality good but risk of bias unknown; trials mostly excluded complicated pregnancies, so it says little about high-risk pregnancy; English full-text only. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Samuel 2021 Randomized control trial evidence for the benefits of massage and relaxation therapy on sleep in cancer survivors-a systematic review J Cancer Surviv · PMID 33269414 · DOI	Systematic review	4 randomised trials of massage and 3 of relaxation therapy in cancer survivors	Massage given by massage therapists Not given in the abstract Not summarised in the abstract	The four massage trials showed improved self-reported sleep and longer sleep episodes on an accelerometer; relaxation therapy trials showed no significant improvement. No pooled effect size given (Not given) Methods, limits and funding Outcomes: Sleep (self-reported questionnaires; accelerometer-recorded long sleep episodes) Limitations: Only four massage trials; no pooled analysis; no certainty rating in the abstract; the review did not test whether non-certified providers get the same result. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Shan 2023 Massage therapy significantly improves cancer-related fatigue in cancer patients: a meta-analysis of randomized controlled trials Support Care Cancer · PMID 37452895 · DOI	Systematic review meta analysis	789 11 randomised trials in people with cancer (389 massage, 400 control); results strongest in breast cancer	Massage therapy (including reflexology and Chinese massage subgroups) Not given in the abstract Subgroup analysis suggested 20 to 40 minutes, twice a week, for 3 to 5 weeks had the largest effects (as reported by the review) · vs Control groups (not detailed in the abstract)	Large pooled reduction in fatigue; five studies described adverse events and only two reported any. Fatigue SMD -1.69 overall; -1.62 in breast cancer; reflexology -2.71; Chinese massage -1.14 (Overall -2.46 to -0.93; breast cancer -2.18 to -1.05) Methods, limits and funding Outcomes: Cancer-related fatigue; adverse events Limitations: Large effects and high heterogeneity are likely; the abstract does not give a certainty rating. Small trials; subgroup findings on duration and frequency are exploratory. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Smith 2018 Massage, reflexology and other manual methods for pain management in labour Cochrane Database Syst Rev · PMID 29589380 · DOI	Systematic review meta analysis	1055 14 randomised trials in women in labour; 10 trials (1,055 women) contributed data to the pooled analysis	Massage (and warm packs and other manual methods, reviewed separately) Not given in the abstract During labour; first stage mostly · vs Usual care, other non-drug methods, music, no treatment	Low-quality evidence that massage reduced pain intensity in the first stage of labour compared with usual care. Very low-quality evidence found no clear difference in length of labour, pain medicine, assisted vaginal birth or caesarean section. Single trials reported less anxiety and more sense of control. Few trials reported on safety. First-stage pain SMD -0.81 (6 trials, 362 women); length of labour MD 20.64 minutes (6 trials, 514 women); caesarean RR 0.75 (6 trials, 514 women) (Pain -1.06 to -0.56; length of labour -58.24 to 99.52; caesarean 0.51 to 1.09) Methods, limits and funding Outcomes: Pain intensity, length of labour, pain medicine use, assisted birth, caesarean section, anxiety, sense of control, satisfaction Limitations: Most trials were at high risk of performance and detection bias; no trials of reflexology; few trials reported on the key outcomes or on safety. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Suh 2022 Case report: Traumatic median neuropathy in the distal forearm after massage therapy Front Neurol · PMID 36212666 · DOI	Case report	1 One 61-year-old man	Massage therapy (two sessions) Not given in the abstract Two sessions	Numbness and pins and needles in the right hand after two sessions; tests located a median nerve lesion in the forearm. After injection and a nerve-pain medicine, more than 80 percent of sensory symptoms had resolved at one month. One person (Not given for a case report) Methods, limits and funding Outcomes: Symptoms and nerve tests Limitations: A single case; cause and effect is the authors' judgement. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Win Myint 2025 Effectiveness of Massage Therapy for Cancer Pain, Quality of Life and Anxiety Levels: A Systematic Review and Meta-Analysis J Clin Nurs · PMID 39558520 · DOI	Systematic review meta analysis	3671 36 randomised trials in people with cancer	Massage therapy (types pooled) Not given in the abstract Not summarised in the abstract (results measured after the intervention)	Massage improved pain, quality of life and anxiety after the intervention, but all outcomes were rated very low certainty and most studies had unclear or high risk of bias. Pain SMD -0.51; quality of life SMD 0.48 (higher-is-better scales); anxiety SMD -0.38 (Pain -0.68 to -0.33; quality of life 0.19 to 0.78; anxiety -0.57 to -0.18) Methods, limits and funding Outcomes: Cancer pain, quality of life, anxiety Limitations: Very low certainty (GRADE) for all outcomes; most studies had unclear or high risk of bias. Types of massage and comparison groups are not detailed in the abstract. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
Yin 2014 Adverse events of massage therapy in pain-related conditions: a systematic review Evid Based Complement Alternat Med · PMID 25197310 · DOI	Systematic review of case reports	40 published reports (2003 to 2013) describing 138 adverse events associated with massage, mostly for pain-related conditions	Massage, including some with spinal manipulation Not given in the abstract Reports describe events after the fact; no set duration · vs No comparison group	Main complications were disc herniation, soft tissue trauma, nerve problems, spinal cord injury and vertebral artery dissection. Spinal manipulation within massage was repeatedly linked to serious events. The authors conclude massage is not free of risk but serious events are low in incidence. 138 events in 40 reports (Not given for a case report) Methods, limits and funding Outcomes: Type and frequency of adverse events Limitations: Case reports cannot give a rate; under-reporting is likely; massage and manipulation are mixed in the reports. Funding and conflicts: No funding or conflict statement in the record read. Read: abstract
s-nccih-massage NCCIH: Massage Therapy: What You Need To Know (page marked Last Updated May 2019) · Source	Health agency n/a			Methods, limits and funding Read: full-text

What we do not know yet

No trial read compared massage with a convincingly blinded sham in every condition, so the share of benefit from touch, attention and expectation is unknown.
Almost every trial was unblinded and relied on self-rated symptoms.
Reviews pooled many styles, so which style, how long a session, or how often is not established.
Follow-up beyond a few months was missing for most outcomes.
How often serious harms occur is unknown; case reports cannot give a rate.
History beyond the NCCIH summary was not read from a source.

How new research is added

New studies are found by a saved PubMed search and placed in a review queue before the page changes.

Lab markers

What changed in blood or saliva tests during trials. A changed number is not the same as feeling or being healthier, and we give no "normal ranges".

Other 7

●

Back pain intensity

Lower than inactive treatment in the short term; weaker over longer periods in trials

A self-rated score of how much the back hurts.

Low certainty

Results and what it means

Why it was measured

To see whether massage lowers pain people feel.

Statistical vs. clinical meaning

A real but modest short-term change by standardised score; not a guarantee for any person.

Safety implication

Increased pain was the most common adverse event.

Study	Result	People	Product
⁴	Pain was lower than with inactive controls (SMD -0.75; 7 studies, 761 people) in sub-acute and chronic pain, in the short term but not the long term.	761 people in 7 trials	Massage

⁴

●

Neck pain intensity

Little to no difference against placebo at up to 12 weeks in trials

A self-rated score on a 0 to 100 scale.

Low certainty

Results and what it means

Why it was measured

To see whether massage lowers neck pain compared with a placebo.

Statistical vs. clinical meaning

The interval includes no effect, so the average gain may be nothing.

Safety implication

Serious adverse events were not reported.

Study	Result	People	Product
⁵	3.43 points better than placebo on a 0 to 100 scale, with a confidence interval from 8.16 better to 1.29 worse.	403 people in 8 trials	Any massage, mostly single sessions

⁵

●

Anxiety scores

Lower than comparison groups straight after massage in trials

Self-rated anxiety questionnaires.

Very low certainty

Results and what it means

Why it was measured

To see whether massage lowers how anxious people feel.

Statistical vs. clinical meaning

Self-rated, unblinded and short-term; certainty is very low in cancer and the pregnancy trials were all at some risk of bias.

Safety implication

None reported in the abstracts.

Study	Result	People	Product
¹⁵	Pooled anxiety lower after massage (SMD -0.38, 95% CI -0.57 to -0.18).	36 trials, 3,671 people with cancer (all outcomes)	Massage, types pooled
⁷	Anxiety lower than with usual care (SMD -0.59, 95% CI -1.06 to -0.12; I2 75 percent).	5 studies in pregnant women	Massage

^{15 7}

●

Fatigue score in people with cancer

Lower than controls in trials

A self-rated score of tiredness.

Low certainty

Results and what it means

Why it was measured

To see whether massage reduces fatigue.

Statistical vs. clinical meaning

A large average difference, but from small unblinded trials.

Safety implication

Few adverse events reported.

Study	Result	People	Product
¹²	Fatigue lower with massage (SMD -1.69, 95% CI -2.46 to -0.93).	789 people in 11 trials	Massage, including reflexology and Chinese massage

¹²

●

Pain in the first stage of labour

Lower than usual care in trials

Self-rated pain scores during early labour.

Low certainty

Results and what it means

Why it was measured

To see whether massage eases labour pain.

Statistical vs. clinical meaning

A real average difference in self-rated pain; no clear change in births by caesarean.

Safety implication

Few trials reported safety.

Study	Result	People	Product
¹³	Pain lower with massage (SMD -0.81, 95% CI -1.06 to -0.56), low quality.	362 women in 6 trials	Massage

¹³

●

Daily weight gain in preterm infants

Higher with massage in trials

Grams gained per day.

Low certainty

Results and what it means

Why it was measured

Weight gain decides when a preterm baby may leave hospital.

Statistical vs. clinical meaning

A measured gain; weight is a stand-in for health.

Safety implication

Not reported in the abstract.

Study	Result	People	Product
⁹	5.07 g/day more (95% CI 2.19 to 7.94).	697 infants in 15 trials	Massage, moderate or light pressure

⁹

●

Muscle soreness rating

Lower than no intervention at 24, 48 and 72 hours in trials

A self-rated score of how sore the muscles feel.

Low certainty

Results and what it means

Why it was measured

To see whether massage eases soreness after exercise.

Statistical vs. clinical meaning

Self-rated and unblinded.

Safety implication

Not reported in the abstract.

Study	Result	People	Product
⁶	Soreness lower overall (SMD -1.16, 95% CI -1.60 to -0.72).	504 participants in 11 articles	Massage after strenuous exercise

⁶

Safety

What studies and official sources say about side effects, medicines and who should check first. Each point shows what kind of evidence it comes from.

Main points to know 4

Massage is generally low risk, but rare serious harms are reported. NCCIH says the risk of harmful effects from massage therapy appears to be low, with rare reports of serious side effects such as a blood clot, nerve injury or bone fracture, often after vigorous massage such as deep tissue massage or in people at higher risk of injury, such as older adults. A 2003 systematic review of case reports concluded that massage is not entirely risk free but serious events are probably rare, and that most were linked to unusual manual techniques or non-therapists, with Swedish massage rarely implicated. A 2014 review of 40 case reports (138 events) listed disc herniation, soft tissue injury, nerve damage, spinal cord injury and vertebral artery dissection, and linked spinal manipulation within massage to serious events. Authoritative guidance and systematic reviews of case reports ^{17 3 16}
Soreness after a session is the most common effect. In the low back pain trials, increased pain was the most common adverse event, reported by 1.5 to 25 percent of participants, and no serious adverse events were reported. In the neck pain review, adverse events such as soreness, sweating and low blood pressure were reported but the certainty was very low and serious events were not available. Systematic review of randomised trials ^{4 5}
Tell the therapist about blood clots, bleeding problems, fractures, cancer and pregnancy. No source read gave a firm rule for each condition. The reports above include blood clot related events (pulmonary embolism), bruising or blood collections (haematoma), nerve damage and fractures. NCCIH says people with cancer may need lighter pressure in sensitive areas and advises talking to a health care provider before starting. A review in pregnancy found only minor and transient side effects in the two trials that reported any, and discusses precautions to prevent pulmonary embolism. Authoritative guidance and reviews ^{17 3 10 2}
Massage should not delay medical care. NCCIH advises not putting off seeing a health care provider about a medical problem, and asking your provider whether massage suits you. Authoritative guidance ¹⁷

Side effects 5

Effect	How often	Severity and recovery	Evidence
Soreness and increased pain after a session Neck pain trials also reported sweating and low blood pressure, but the certainty is very low.	Increased pain intensity was reported by 1.5 to 25 percent of participants in back pain trials	Mild Usually temporary; the reviews do not give durations	Systematic review of randomised trials ^{4 5}
Nerve injury Reported after two massage sessions in the case; the authors say massage should be done carefully.	Rare; counted in case reports, no rate available	Can be lasting In one case report of a 61-year-old man, more than 80 percent of the numbness and tingling had resolved a month after treatment	Case reports and reviews of case reports ^{14 3 16 17}
Fracture or spinal injury One case report describes a neck fracture-dislocation with loss of nerve function after traditional massage in a man whose ankylosing spondylitis (a stiffening spine disease) had not been diagnosed. NCCIH mentions bone fracture among rare serious effects.	Rare; case reports only	Can be very serious Can be permanent	Case reports ^{1 16 17}
Blood clot, bleeding or bruising events The 2003 review lists pulmonary embolism, haematoma, pseudoaneurysm and leg ulcers among reported events. It does not say which people are at risk.	Rare; case reports only	Can be serious Varies	Systematic review of case reports ^{3 17}
Neck artery dissection and stroke The reviews list vertebral artery dissection and cerebrovascular accidents. The 2014 review links serious events to spinal manipulation within massage, so this belongs mainly to neck manipulation and forceful neck techniques, not to gentle massage.	Rare; case reports only	Can be very serious Varies	Systematic reviews of case reports ^{16 3}

Liver

No source read links massage to liver problems.

This is a safety connection. It is not evidence that Massage Therapy supports liver health.

Medicines and supplements 2

With	Why it matters	Importance	Evidence	What to do
Blood-thinning medicines and bleeding disorders prescription	Theory: deep pressure could cause bruising or bleeding under the skin. Not tested in the sources read.	No interaction study was found. The case-report review lists haematoma among reported harms but does not say whether those people took blood thinners.	Absence of evidence Not enough evidence ³	Absence of a study is not proof of safety. People who take blood thinners or have a bleeding disorder should ask their doctor before deep or vigorous massage.
Cancer treatment prescription	Not described in the sources read.	NCCIH says massage can be part of supportive care with appropriate precautions, and that therapists may need to use less pressure on sensitive areas. A 2005 review says people with cancer may be at higher risk of rare adverse events.	Authoritative guidance Not enough evidence ^{17 2}	Tell your cancer care team and your therapist before starting.

Who should check first 7

Situation	Advice	Evidence
Blood clots (deep vein thrombosis) or a history of them	No source read gives a rule, but NCCIH and the case-report reviews list blood clot related events, including pulmonary embolism, after massage. Anyone with a known or suspected clot, or a swollen, painful, warm leg, should get medical advice before any deep leg massage.	Case reports and authoritative guidance ^{17 3}
Bleeding disorders and blood-thinning medicines	No source read addressed this directly. Haematoma is among reported harms. Ask a doctor before deep or vigorous massage.	Absence of evidence ³
Cancer, including tumour sites	NCCIH says massage can be part of supportive care with appropriate precautions and that therapists may need lighter pressure in sensitive areas. No source read says exactly where on the body to avoid; this needs the cancer care team.	Authoritative guidance ^{17 2}
Fractures, osteoporosis and fragile spine	NCCIH lists bone fracture among rare serious effects, often after vigorous massage or in people at higher injury risk such as older adults. A case report describes a neck fracture-dislocation in a man with undiagnosed ankylosing spondylitis. No source read gave fracture rates in osteoporosis.	Authoritative guidance and case report ^{17 1}
Pregnancy	A review of 12 trials of relaxation massage in healthy pregnant women (at least 12 weeks along) found only minor, transient side effects in the two trials that reported any, and discusses precautions to prevent pulmonary embolism. Seven trials excluded complicated pregnancies, so it says little about them. NCCIH's page does not address pregnancy. A midwife or doctor should be asked first.	Systematic review ^{10 17}
Skin infections, open wounds and fever	No source read addressed these. This page cannot cite a source for rules on massage over broken or infected skin or during fever, so ask the therapist or a doctor.	Absence of evidence
Older adults and anyone at higher injury risk	NCCIH says serious effects often involved vigorous massage or people at higher injury risk, such as older adults; discuss concerns with a health care provider before starting.	Authoritative guidance ¹⁷

Overdose, product quality and regulators

Summary. Massage is a hands-on practice, so there is no toxic dose. The documented harms are injuries from pressure or forceful technique, listed under adverse effects.

Human doses studied. Session lengths and numbers used in trials are given where reported in the Evidence tab. They describe what was studied, not what anyone should do.

High dose or long term signals. A neck pain review found that trials with more and longer sessions looked better than single sessions, and no source read measured harm from very frequent massage.

The nearest quality issue in the sources is who gives the massage and how they were trained.

Qualified practitioners. NCCIH says 45 US states and the District of Columbia regulate massage and require therapists to be licensed or certified, usually after an approved programme and an exam; some also hold national certification from the National Certification Board for Therapeutic Massage & Bodywork. It advises asking about training, experience, credentials, the number of treatments expected, cost and insurance. The 2003 review found most serious events came from unusual techniques or non-therapists, while therapists were rarely implicated. ^{17 3}

Body	Action	Date
National Center for Complementary and Integrative Health (NCCIH, part of the US National Institutes of Health) (United States)	States that the risk of harmful effects from massage therapy appears to be low, with rare reports of serious effects such as a blood clot, nerve injury or bone fracture; notes that cancer patients may need modified pressure; describes licensing in 45 states and DC; advises against delaying medical care. source	2019-05

When to get medical help

Sudden severe neck or back pain, or new numbness, tingling or weakness in an arm or leg during or after a session: stop and get medical advice (reports describe nerve injury and spinal injury).
Sudden chest pain, trouble breathing or coughing blood after massage: call emergency services (pulmonary embolism is among the reported harms).
A leg that becomes swollen, painful, warm or red, especially one side: get medical advice before any further leg massage.
Sudden severe headache, dizziness, trouble speaking or face weakness after neck work: call emergency services (stroke and artery dissection are reported with neck techniques).
Large unexplained bruising, a growing swelling, or pain that keeps getting worse over days: get medical advice.

Stop using the product and contact a clinician or emergency service. In the US, Poison Control is 1-800-222-1222.

General information only, not medical advice. Ask a clinician or pharmacist before use, especially with medicines, in pregnancy or with a health condition. See the Natural Remedy Safety guide.

Frequently asked questions

What is massage therapy?

NCCIH describes it as manipulating the body's soft tissues to help manage a health condition or enhance wellness. It has been practised in most cultures throughout history.

¹⁷

Does massage help low back pain?

A Cochrane review of 25 trials found short-term pain relief compared with inactive treatment, but not at longer follow-up. The authors have very little confidence in the result because the evidence is low to very low quality.

⁴

Does massage help neck pain?

A 2024 Cochrane review of 33 trials found low-certainty evidence of little to no difference against placebo. Trials with more and longer sessions looked better.

⁵

Does massage help anxiety?

Pooled trials in people with cancer and in pregnant women found lower anxiety scores with massage, but the certainty was very low or the trials were biased. It is not a proven treatment for anxiety.

^{15 7}

Is massage helpful during cancer treatment?

A 2025 analysis of 36 trials found less pain and anxiety and better quality of life, all very low certainty. NCCIH says it can be part of supportive care with precautions such as lighter pressure in sensitive areas. Talk to your care team first.

^{15 12 17}

Does massage help fibromyalgia?

Massage lasting 5 weeks or more eased pain, anxiety and depression straight afterwards in 9 small trials, but not sleep. No follow-up data were available.

⁸

Does massage ease labour pain?

The Cochrane review found lower first-stage labour pain, at low quality, and no clear difference in length of labour or caesarean births. Few trials reported safety.

¹³

Does massage help sore muscles after exercise?

Pooled trials found lower soreness at 24, 48 and 72 hours compared with no intervention, plus higher muscle force and lower creatine kinase.

⁶

Which massage style is best?

No review read compared styles head to head. NCCIH lists Swedish, sports and clinical massage, but the evidence cannot say which is best.

^{4 17}

Is massage safe?

NCCIH says the risk of harm appears to be low, with rare reports of blood clots, nerve injury or fracture. Soreness is the most common effect. Reviews of case reports list rare serious harms, mostly with unusual techniques, neck manipulation or non-therapists.

^{17 4 3 16}

Who should be careful with massage?

People with blood clots, bleeding disorders or blood thinners, cancer, pregnancy, fractures or osteoporosis, and older adults, should ask a doctor first. No source read gave firm rules for each, so personal medical advice matters.

^{17 3 10}

Can massage replace medical treatment?

No. NCCIH says not to delay seeing a health care provider about a medical problem, and to ask your provider whether massage suits you.

¹⁷

Explore by health goal

Pain Relief Anxiety Support Calm Relaxation Mood Balance Better Sleep Reduce Fatigue Womens Wellness Muscle Recovery

Editorial transparency

Written by: AlternativeMed editorial team. Research compiled with AI assistance, then checked against PubMed records.
Medical review: None. This page has not been reviewed by a physician or pharmacist.
First published: 2026-10-31
Last substantive update: 2026-10-31
How we research: We searched PubMed, re-read the abstract of every review and trial cited, and read the NCCIH page on massage therapy through a tool that summarises pages. Certainty ratings are the compiler's own reading, except where a review gives its own GRADE (stated in the text). Funding is not stated in most abstracts, and this is noted where relevant.
How we rate evidence: See "How we rate evidence" in the Evidence tab.
Independence: No product is promoted on this page. No affiliate links are shown.

Sources 17

Each source was opened and checked. Where we read only the abstract, it says so. The small numbers in the text point here.

Abilash K, Mohd Q, Ahmad Z. Fracture-dislocation at C6-C7 level with Quadriplegia after Traditional Massage in a Patient with Ankylosing Spondylitis: A Case Report. Malays Orthop J. 2017. PMID 29021885. DOI 10.5704/MOJ.1707.013. (abstract)
Corbin L. Safety and efficacy of massage therapy for patients with cancer. Cancer Control. 2005. PMID 16062163. DOI 10.1177/107327480501200303. (abstract)
Ernst E. The safety of massage therapy. Rheumatology (Oxford). 2003. PMID 12777645. DOI 10.1093/rheumatology/keg306. (abstract)
Furlan AD, Giraldo M, Baskwill A et al.. Massage for low-back pain. Cochrane Database Syst Rev. 2015. PMID 26329399. DOI 10.1002/14651858.CD001929.pub3. (abstract)
Gross AR, Lee H, Ezzo J et al.. Massage for neck pain. Cochrane Database Syst Rev. 2024. PMID 38415786. DOI 10.1002/14651858.CD004871.pub5. (abstract)
Guo J, Li L, Gong Y et al.. Massage Alleviates Delayed Onset Muscle Soreness after Strenuous Exercise: A Systematic Review and Meta-Analysis. Front Physiol. 2017. PMID 29021762. DOI 10.3389/fphys.2017.00747. (abstract)
Hall HG, Cant R, Munk N et al.. The effectiveness of massage for reducing pregnant women's anxiety and depression; systematic review and meta-analysis. Midwifery. 2020. PMID 32827841. DOI 10.1016/j.midw.2020.102818. (abstract)
Li YH, Wang FY, Feng CQ et al.. Massage therapy for fibromyalgia: a systematic review and meta-analysis of randomized controlled trials. PLoS One. 2014. PMID 24586677. DOI 10.1371/journal.pone.0089304. (abstract)
Lu LC, Lan SH, Hsieh YP et al.. Massage therapy for weight gain in preterm neonates: A systematic review and meta-analysis of randomized controlled trials. Complement Ther Clin Pract. 2020. PMID 32379694. DOI 10.1016/j.ctcp.2020.101168. (abstract)
Mueller SM, Grunwald M. Effects, Side Effects and Contraindications of Relaxation Massage during Pregnancy: A Systematic Review of Randomized Controlled Trials. J Clin Med. 2021. PMID 34441781. DOI 10.3390/jcm10163485. (abstract)
Samuel SR, Gururaj R, Kumar KV et al.. Randomized control trial evidence for the benefits of massage and relaxation therapy on sleep in cancer survivors-a systematic review. J Cancer Surviv. 2021. PMID 33269414. DOI 10.1007/s11764-020-00972-x. (abstract)
Shan S, Lin L, Fang Q et al.. Massage therapy significantly improves cancer-related fatigue in cancer patients: a meta-analysis of randomized controlled trials. Support Care Cancer. 2023. PMID 37452895. DOI 10.1007/s00520-023-07926-w. (abstract)
Smith CA, Levett KM, Collins CT et al.. Massage, reflexology and other manual methods for pain management in labour. Cochrane Database Syst Rev. 2018. PMID 29589380. DOI 10.1002/14651858.CD009290.pub3. (abstract)
Suh MR, Song JH, Park Y et al.. Case report: Traumatic median neuropathy in the distal forearm after massage therapy. Front Neurol. 2022. PMID 36212666. DOI 10.3389/fneur.2022.959919. (abstract)
Win Myint O, Yoong SQ, Toh E et al.. Effectiveness of Massage Therapy for Cancer Pain, Quality of Life and Anxiety Levels: A Systematic Review and Meta-Analysis. J Clin Nurs. 2025. PMID 39558520. DOI 10.1111/jocn.17547. (abstract)
Yin P, Gao N, Wu J et al.. Adverse events of massage therapy in pain-related conditions: a systematic review. Evid Based Complement Alternat Med. 2014. PMID 25197310. DOI 10.1155/2014/480956. (abstract)
. NCCIH: Massage Therapy: What You Need To Know (page marked Last Updated May 2019). . . (full-text)

Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.