The first-choice lower-back pain killer, paracetamol, worked no better than dummy drugs administered in a trial of more than 1,600 people suffering from the condition, researchers said.
In fact, the median recovery time for those on placebo was a day shorter than that for trial subjects given real medicine, they wrote in The Lancet medical journal.
"Our findings suggest that... paracetamol does not affect recovery time compared with placebo in low-back pain, and question the universal endorsement of paracetamol in this patient group," the Australian team concluded.
Lower-back pain is the leading cause of disability in the world, and paracetamol is "universally" recommended as the treatment of first choice, said a statement carried by The Lancet.
The Paracetamol for Low-Back Pain Study (PACE) divided 1,652 individuals with acute pain from 235 clinics in Sydney, Australia, into three trial groups.
One group received regular paracetamol doses, the other used the drug as needed, and the third was given placebo pills.
Recovery was defined as seven consecutive days of 0 or 1 pain intensity on a 0-10 scale.
"Median time to recovery was 17 days in the regular paracetamol group, 17 days in the as-needed paracetamol group, and 16 days in the placebo group," said the statement.
All patients were given high-quality advise and reassurance, and the findings suggest these may be more important in lower-back pain management than drug therapy, said the authors.
"Our results convey the need to reconsider the universal endorsement of paracetamol in clinical practice guidelines as first-line care for low-back pain."
A potential limitation of the study was that some participants used other treatments.
In a comment also carried by The Lancet, Bart Koes and Wendy Enthoven from the Universal Medical Center in Rotterdam applauded the team "for tackling this research question on a topic that has been without debate and evidence for such a long time."
But they cautioned that guidelines should not be changed on the basis of a single trial.