Spinal manipulation - which is used by chiropractors and osteopaths in the UK to treat neck and back pain - is of little help, researchers have said.
Back pain is one of the conditions treated with spinal manipulation
Experts from Peninsula Medical School in Devon reviewed 26 studies carried out between 2000 and 2005.
Writing in Journal of the Royal Society of Medicine, they said the data gave "little evidence" of effectiveness.
Chiropractors said the team had focused on negative studies which supported the researchers' views - a claim they deny.
The researchers said they looked at all studies evaluating the benefits of spinal manipulation for period pain, colic, asthma, allergy and dizziness - as well as back and neck pain up to 2005.
It was found the data did not show spinal manipulation was effective for any condition - except for back pain where it is superior to sham manipulation, but not better than conventional treatments.
The researchers said that, as spinal manipulation had been linked to mild side effects in around half of patients, such as temporary stiffness, and - much more rarely - strokes brought on by damage to the vertebral artery in the back, it was not something which should be used instead of other therapies.
They suggest existing guidelines need to be re-evaluated in the light of their conclusions.
Professor Edzard Ernst, who led the review, said: "There is little evidence that spinal manipulation is effective in the treatment of any medical condition.
"The findings are of concern because chiropractors and osteopaths are regulated by statute in the UK.
"Patients and the public at large perceive regulation as proof of the usefulness of treatment.
"Yet the findings presented here show a gap and contradiction between the effectiveness of intervention and the evidence."
"We suggest that the guidelines be reconsidered in the light of the best available data."
Professor Ernst said the findings should be seen as a "wake-up call" to the chiropractic profession.
"One way forward is more rigorous clinical trials to test the efficacy of spinal manipulation," he added.
"After all, the treatment is not without risk and chiropractors must demonstrate why it should be a recommendable medical treatment option."
But in a statement, the British Chiropractic Association said it was disappointed by the study's conclusions, which it believed were based on "negative" research - other studies had come to the opposite conclusion.
"The usefulness of manipulation is that it can be added, substituted or modified as part of a package of care that provides management, pain control, advice and recognises risks to a good recovery," it said.
"Recent clinical trials funded by the Medical Research Council show that manipulation is effective and cost-effective within such a package for back pain."
The National Council for Osteopathic Research accused Professor Ernst of working with out of date data.