Shiatsu is a manual therapy from Japan in which the practitioner presses on the body with the thumbs, fingers and palms, leaning in with their own body weight. Its name means finger pressure. A 2011 systematic review by Robinson, Lorenc and Liao at London South Bank University describes it as using pressure, stretches and gentle manipulation from the fingers, thumbs, elbows, knees and feet over the whole body, with diagnosis made mainly by touch rather than by the pulse and tongue of Chinese medicine. That review found too few studies to judge shiatsu, and the trials published since are small and rarely compare it with a sham.
Where does shiatsu come from?
A 2018 paper by Cabo and colleagues, written from the Shiatsu Research Institute in London, says shiatsu has been taught in specialised schools in Japan since 1940 and was recognised in 1964 by Japan’s Ministry of Health as a Japanese therapeutic treatment. It names three practitioners whose work underlies almost every style taught worldwide: Tokujiro Namikoshi, Shizuto Masunaga and Katsusuke Serizawa. Namikoshi’s style, the paper says, is the one most widely practised in Japan. The Robinson review lists the styles most often found in Britain, among them Zen shiatsu, Namikoshi shiatsu, Macrobiotic shiatsu and Hara shiatsu, and notes that they differ in their theory as well as their technique.
Who can practise shiatsu in Japan?
In Japan shiatsu is a licensed occupation. Article 1 of the Act on Practitioners of Massage, Shiatsu, Acupuncture and Moxibustion (Act No. 217 of 1947) requires anyone other than a physician who practises anma, massage or shiatsu as a business to hold an anma-massage-shiatsu licence. Under Article 2, the Minister of Health, Labour and Welfare grants it to people who have studied anatomy, physiology, pathology, hygiene and the other knowledge and skills the work requires for at least three years at an accredited school or training facility and then passed a national examination. The same Act issues separate licences for acupuncture and for moxibustion; Japanese acupuncture covers those.
How is shiatsu different from acupressure and tuina?
Robinson and colleagues treated shiatsu and acupressure together. Their review describes shiatsu as incorporating acupressure, which they define as similar pressure held longer on specific points chosen by the theory of Chinese medicine, and it argues that evidence for acupressure may support claims for shiatsu. Cabo and colleagues reject that step. They say shiatsu pressure is always stationary and sustained, with the thumb kept straight and the force coming from the practitioner’s body weight, while acupressure often uses circling or pumping pressure with the thumb bent and the strength of the arm or hand. Shiatsu treats most of the body in a session; acupressure, in most cases, treats a few points. Of the 9 shiatsu studies in the 2011 review, they found only one that used shiatsu alone and described how it was given. Their conclusion is that acupressure trials should not be counted as evidence for shiatsu. Acupressure reports those trials on their own terms.
Tuina is the manual therapy of Chinese medicine. The WHO benchmarks for the practice of tuina (2020) list eight major manipulations, among them rolling, kneading, rubbing and grasping, and allow a medium such as a liniment or ointment on the skin; Cabo and colleagues note that shiatsu normally uses no oil or cream. Tuina sets out those manipulations and the trials of them.
What did the 2011 review find?
Robinson and colleagues searched 13 databases for trials published after January 1990 and screened 1,714 publications. They included 9 studies of shiatsu and 71 of acupressure. The shiatsu studies were 1 randomised trial, 3 controlled but non-randomised studies, 1 within-subjects trial, 1 observational study and 3 uncontrolled studies, each on a different condition, from chronic stress and schizophrenia to fibromyalgia and inducing labour, so none could be pooled. The single randomised trial gave shiatsu inside a package of integrated care for back and neck pain, against standard care, in 80 people, and found no significant difference. The authors judged the shiatsu evidence insufficient in both quantity and quality, and graded most of the shiatsu studies C, their lowest grade.
What have trials found since 2011?
| Trial | Who and how many | Comparison | What it reported |
|---|---|---|---|
| Kobayashi 2019, Tokyo | 59 adults with low back pain for over 12 weeks | Weekly one-hour shiatsu for 4 weeks added to standard care, against standard care alone; outcome assessors blinded | No significant difference on the main disability score at week 4 or week 8 (1.7 points at week 8, p = 0.08); better scores at week 8 on present pain, the Oswestry Disability Index and quality of life |
| Kobayashi 2023, same trial | The same 59 people | Secondary analysis of sleep | Better Pittsburgh Sleep Quality Index scores at weeks 4 and 8; use of sleep medication worse in the shiatsu group in several models |
| Yuan 2013 | 34 adults with fibromyalgia | Shiatsu twice a week for 8 weeks, against an educational booklet | Larger changes in pain, pressure pain threshold, sleep and fibromyalgia impact; none in anxiety; no adverse effects reported |
| Villani 2017, Rome | 37 people with headache that had not responded to 2 or more preventive drugs | Shiatsu, amitriptyline, or both, for 3 months | No difference between groups in the share with headache days halved; fewer painkillers in the shiatsu groups; all 7 adverse events attributed to amitriptyline |
| Lanza 2018 | People with Alzheimer’s disease and depression | Weekly shiatsu for 10 months plus physical activity, against physical activity alone | A fall in depression scores in the shiatsu group only |
| Harorani 2021 | 68 patients on mechanical ventilation | Three 5-minute periods of shiatsu, against a touch on the same area | Lower agitation scores after the intervention |
| Teimoori 2014, Zahedan, Iran | 288 women past their due date | Pressure on three points by a midwife, against routine care | Spontaneous labour in 82 women (56.9%) in the shiatsu group and 12 (8.3%) in the control group |
| Norhapifah 2024, Samarinda, Indonesia | 80 women in their first labour, at four clinics | Shiatsu by midwives with routine care, against routine care; two clinics formed each group | Pain scores fell from 6.85 to 4.78 out of 10 by transition with shiatsu and rose to 8.85 without; anxiety moved the same way |
Only Harorani gave the control group a touch to match the treatment; the others compared shiatsu with no added treatment, a booklet, a drug or routine care, so in none of them could participants be kept unaware of whether they had received shiatsu. Teimoori’s three points were GB21 (Jianjing), LI04 (Hegu) and SP06 (Sanyinjiao). AcuiQ’s catalogue records all three as traditionally forbidden in pregnancy and as points used to bring on labour at term; Pregnancy and the forbidden points explains how both readings sit in one entry. The Indonesian trial allocated whole clinics rather than individual women. The 2018 Cochrane review of manual methods for labour pain, by Smith and colleagues, listed shiatsu among the manual methods it searched for and left acupressure to a separate Cochrane review, which is the line Cabo and colleagues draw.
Have later reviews pooled shiatsu?
A 2015 review of massage styles for fibromyalgia by Yuan, Matsutani and Marques included 10 trials of several styles; its pooled analysis reported on myofascial release, and it summarised the shiatsu evidence narratively and rated it limited. A 2026 review of touch-based therapies in dementia care by Mitchell and colleagues included 33 studies, of which 2 were of reflexology or shiatsu. It found short-term calming the most consistent result, clearest for massage and acupressure, rated certainty low and did not meta-analyse. A Europe PMC search on 13 October 2026 returned no systematic review since 2011 confined to shiatsu.
Is shiatsu safe?
The fibromyalgia, headache and labour trials above reported no adverse effects of shiatsu, but none was large enough to detect an uncommon one. A 2012 case report describes a 76-year-old man whose stents in both common iliac arteries were found collapsed five months after they were placed; he had had shiatsu on the area around the navel every day. He had no symptoms, and the authors advise patients with iliac stents to avoid pressure on the abdominal wall.
What the reviews could not settle
No review has yet pooled randomised trials of shiatsu for any condition, because there are too few of them on any one condition and they use different protocols, comparisons and outcomes. Whether the acupressure literature can stand in for shiatsu is itself disputed between the 2011 review and Cabo and colleagues, and no trial has compared the two techniques directly. The trials since 2011 rarely use a sham, so none separates the effect of shiatsu from that of attention and touch. A protocol published in 2023 by Bernardinelli and colleagues describes a crossover trial of 120 people with axial spondyloarthritis in Orléans, France, comparing three active and three sham shiatsu sessions for fatigue, and is registered as NCT05433168.