It has found paracetamol, which is currently recommended as a first-line treatment for lower back pain and is one of the most widely used over-the-counter pain medications in the world, may be virtually useless for the condition.

The same analysis, which is published Wednesday in the British Medical Journal and includes data from thousands of sufferers, also found paracetamol provided only minimal benefits for osteoarthritis of the hip or knee.

The senior author of the study, Manuela Ferreira, said the reduction in pain seen for osteoarthritis patients was so small that "we question whether it offers clinical benefit, given the risks and costs of the treatment".

Associate Professor Ferreira said despite years of study, scientists the world over had never been able to definitively explain what causes the sometimes crippling condition.

"There are many factors involved in the pathogenesis of lower back pain," said the George Institute for Global Health and University of Sydney researcher. "But there is more going on than just pain that can be handled by drugs."

It is thought up to 80 per cent of Australians experience lower back pain during their lives, with at least 10 per cent of people experiencing significant disability.

Associate Professor Ferreira said her research, which examined 13 high-quality trials into paracetamol and pain, found there was no evidence it improves pain intensity or quality of life in the short term.

She said there could be many different contributing factors for back pain, complicating treatment, but in the short term the best advice was to encourage people to continue with their daily lives and daily exercise.

In the long-term, eating healthily, doing exercise, and cutting workplace stress were known to help.

"We are just coming to that point that we see the simple solutions are not as simple as we used to think, and we need to get on top of these other issues, particularly stress levels," she said.

Study co-author Ric Day, an expert in pharmacology and arthritis from St Vincent's Hospital and the University of NSW, said whereas evidence on the benefits of treatments such as exercises was growing, it was decreasing for medications.

"We need other approaches, really, and we do have them, they are just not as easy as taking a pill," he said.

Professor Day said that the problem was not that paracetamol didn't relieve pain, it was just that it didn't appear particularly helpful for this type of pain.

"You can only presume that the mix of physical pathology causing the pain in the conditions we are talking about doesn't have enough of a paracetamol-sensitive mechanism to make a substantive difference," he said.

Michael Yelland, and Associate Professor from Griffith University and a fellow of the Royal Australian College of General Practitioners, said people should not be alarmed that the study also identified some abnormal liver function in the arthritis patients taking paracetamol, as if used according to the recommended dose it was safe.

He said it was also worth noting the research only showed paracetamol had no better effect than a placebo, not that it had no effect at all.

"Placebo does not mean 'no effect', people respond to placebos as well, so some people might get a response to paracetamol … people get some effect from just doing something," he said.

Rachelle Buchbinder, a Monash University professor and head of the Cochrane Musculoskeletal group said back pain was an area rife with over-treatment.

"You can get a treatment in the acute phase and it will appear as if it was the treatment that worked, but actually the pain resolved because that is the natural course of the condition," she said.

"[In addition] As we age normally we will see lots of abnormalities on CT scans, MRIs and X-rays but they often have very little to do with what is causing the pain, but when people see the imaging they worry and that might lead to further treatments and investigations they don't need," she said.