Osteopathy

Origins in the United States

The World Health Organization (WHO) benchmarks document on osteopathy training states that the practice was developed in the mid-1800s by Andrew Taylor Still, a physician and surgeon in the United States, who established the first independent school of osteopathy in 1892. The document was prepared with osteopathic practitioners, experts and regulators, so it is a statement from within the profession and not an independent historical study. The sources reviewed for this page do not give a verified date for Still's first use of the term.

Separate paths in the United States and Britain

An article in The DO, a publication of the American Osteopathic Association (AOA), says that John Martin Littlejohn became dean of Still's school in Kirksville, Missouri, in 1899, left within months after a dispute over whether osteopathy should be integrated into conventional medical training, and helped found a college in Chicago in 1900. It says that the 1910 Flexner report on medical schools pushed American osteopathy toward a full medical training model, and that Littlejohn established the British School of Osteopathy in London in 1917.

The same article says that British trainees were denied hospital access and dissection training, so British osteopathy began as a manipulation-only discipline outside medical licensure, with a voluntary register opening in 1939. It also states that only in the United States are graduates of osteopathic medical schools fully licensed physicians, and that elsewhere graduates typically perform only manipulation. This is an account from a professional association and has not been checked against independent histories.

One name, different professional meanings

The WHO document says osteopathy is also called osteopathic medicine and that it relies on manual contact for diagnosis and treatment. In the United States, the AOA describes a Doctor of Osteopathic Medicine (DO) as a physician who practices in all medical specialties, and says DOs make up approximately 11% of physicians in the country. In other countries, the DO article says, osteopathy graduates typically perform only manipulation, so the same word can describe a physician or a manual therapist depending on jurisdiction.

Techniques and overlapping labels

The WHO document uses osteopathic manipulative treatment (OMT) for an array of manual techniques that may be combined with advice on diet, activity, posture or counseling. It groups techniques as direct, indirect, combined, fluid or reflex-based, and says direct techniques include muscle energy, thrust and articulatory maneuvers. It describes osteopathy as distinct from physiotherapy and chiropractic despite some overlap in techniques, and notes that in some countries manual therapists use osteopathic techniques without proper training.

Research papers also use osteopathy as an umbrella. A review of chronic low back pain counted osteopathic manipulative treatment, myofascial release, craniosacral treatment and visceral manipulation as osteopathic interventions, and a separate meta-analysis describes craniosacral therapy as a widely taught and used component of osteopathic medicine. Pooled results for one technique should not be transferred to another.

Assessment and treatment by hand

According to the WHO document, practitioners use manual contact for both diagnosis and treatment and describe a relationship between body structure and function. They use the term somatic dysfunction for impaired or altered function of the body's framework and related vascular, lymphatic and neural elements, and describe manual techniques as aimed at supporting physiological function and the body's self-regulating tendencies. These are concepts of the osteopathic tradition, and the sources reviewed here do not present them as measured biomedical findings.

The National Health Service (NHS) describes osteopathy in the United Kingdom as using massaging and stretching to treat injuries, improve movement and relieve pain. The page reviewed gives no detail on session length, number of visits or typical sequence, so those details are not covered here. A person can ask a practitioner which techniques will be used and can ask for treatment to stop.

Training routes

The WHO benchmarks describe two training types. Type I programs are for people with little or no prior health care training and are typically four-year, full-time programs; a typical Type I program is described as 4,200 hours with at least 1,000 hours of supervised clinical practice. Type II programs are for people who already hold a health care qualification and typically run about 1,000 hours. The document is a reference for national authorities and has no force by itself.

In the United Kingdom, the General Osteopathic Council (GOsC) describes itself as the regulator for osteopaths. It says it keeps a public register, sets education and training standards, and protects the osteopathic title. In the United States, the AOA describes DOs as physicians who receive special training in the musculoskeletal system. Requirements differ by country, and this page does not review legislation.

How the approach is described

The WHO document states that osteopathy is not limited to the diagnosis and treatment of musculoskeletal problems, and that practitioners consider neuromusculoskeletal findings in relation to the overall health of the person. The NHS page describes treatment of injuries, movement and pain. The sources reviewed here do not give figures for why people attend or which complaints are most common.

Conditions that have been studied

An overview of 27 systematic reviews covered nine therapeutic indications, including acute and chronic neck pain, low back pain, pain in pregnancy, chronic non-oncologic pain, migraine, irritable bowel syndrome, fibromyalgia, vertigo, and conditions in premature infants and children. Other reviews have examined osteopathic treatment for infant gastrointestinal disorders such as colic. That a condition has been studied does not mean the research supports a benefit for it.

Overview of systematic reviews

A 2025 overview of 27 systematic reviews of randomized trials reported moderate-quality evidence that osteopathic treatment reduced pain and partly improved physical function in adults with acute neck pain, chronic non-specific low back pain, low back pain in pregnancy and chronic non-oncologic pain, and that it might shorten hospital stays in premature infants. It found no relief for acute non-specific low back pain, fibromyalgia or headaches of various causes, and insufficient evidence for vertigo, chronic obstructive pulmonary disease and hypertension. The authors noted that the underlying studies were heterogeneous and partly of poor quality. Only the abstract of this overview was available for this page.

Low back and neck pain: results depend on the comparator

A 2024 review of nine studies of osteopathic manipulative treatment based on a diagnosis of somatic dysfunction compared it with sham or placebo for neck or low back pain. The pooled differences were not statistically significant for pain intensity (standardized mean difference -0.15, 95% confidence interval -0.38 to 0.08; seven studies, 1,173 patients) or disability (-0.09, -0.25 to 0.08; six studies), and the certainty of evidence was moderate to very low. The authors concluded it was not superior to sham.

A 2026 review of seven trials in chronic low back pain (about 378 participants) combined sham or attenuated treatment, usual care and wait-list comparators and reported a 1.32-point reduction in pain on a 0 to 10 scale at six weeks or less (95% confidence interval -1.71 to -0.93), with moderate certainty. A 2020 review of ten trials, none judged at low risk of bias in every domain, reported larger pooled effects against control interventions. A 2022 review in non-specific neck pain rated its evidence very low. Findings therefore differ with the comparator and the review method.

Cranial techniques and infants

A 2024 meta-analysis of 24 randomized trials (1,613 participants) of craniosacral techniques found no significant effects when analyzed by primary outcome, and its authors cautioned that the few significant secondary results had wide prediction intervals and high bias. In infants, a 2022 review of nine studies found a reduction in crying hours in two trials of infantile colic (moderate certainty), but the authors said conflicting evidence and a lack of high-quality replication prevented generalization.

Trial quality and blinding

Several reviews report that included trials were at some risk of bias, with none judged at low risk in every domain in the chronic low back pain and neck pain reviews, and with certainty ratings of low or very low for many outcomes. The reviews reported different conclusions on similar questions depending on how sham treatment was defined and which comparators were pooled. The abstracts read for this page do not describe blinding of practitioners or participants, so that limitation is not assessed here.

Authors' affiliations

The PubMed records list the authors of several reviews cited here as affiliated with osteopathic colleges or osteopathy research departments, including Lake Erie College of Osteopathic Medicine, the SOMA osteopathy institute in Milan, and Noorda College of Osteopathic Medicine. This does not show bias, and some of these reviews reached null or low-certainty conclusions, but much of the review literature is produced within the field being assessed.

Reported adverse events

A systematic review of adverse events after manual therapies delivered by regulated therapists, which covers manual therapy in general and not osteopathy alone, included eight prospective cohort studies and 31 randomized trials. It estimated minor or moderate transient adverse events at about 41% (95% confidence interval 17 to 68%) in the cohort studies and 22% (11.1 to 36.2%) in the trials, and major adverse events at about 0.13%. The authors concluded that the risk of major events is low but that about half of patients may have minor to moderate reactions.

Reporting in osteopathy trials is patchy. A review of dropouts in 11 trials in chronic pain (1,015 participants) found no significant difference in dropout compared with controls but noted that most articles did not explicitly report adverse events. The 2026 low back pain review likewise said adverse events were infrequently reported, with no serious events attributed to treatment where monitored. An absence of reports does not show that none occurred.

Contraindications and care boundaries

The WHO document lists conditions that osteopathic practitioners treat as contraindications. For direct techniques these include a suspected bleeding disorder, anticoagulant therapy without recent evaluation, compromised bone or joint integrity, aortic aneurysm, suspected vertebral artery compromise, acute fracture, acute cauda equina syndrome and neoplasm at the treated site, with osteoporosis and osteopenia as relative contraindications. The document states that little published evidence exists on which techniques to avoid, and that the lists rest on practitioners' reasoning about biological plausibility.

The NHS page says osteopathy may not be suitable for people who take blood-thinning medicines, have an increased risk of bleeding or stroke, have weak or seriously damaged bones or joints, or have a nerve condition. It advises telling any therapist about medical conditions and medicines, and seeing a general practitioner for symptoms that persist or keep returning. Osteopathy is not a substitute for assessment or emergency care, and new weakness, numbness or loss of bladder or bowel control needs prompt medical assessment.

Training and registration

Training and registration differ by country. The WHO document notes that some manual therapists use osteopathic techniques without proper training, and the NHS page recommends checking a therapist's registration with the General Osteopathic Council. Because the same title covers physicians in the United States and manual therapists elsewhere, a person can ask about a practitioner's qualification, scope and registration.