Most of the antidepressants prescribed for chronic pain are not backed up by evidence, a Cochrane review has found.

The review, which looked at 176 randomised controlled trials including almost 30,000 patients, concluded that there is “insufficient evidence” the commonly prescribed drugs amitriptyline, fluoxetine, citalopram, paroxetine and sertraline are effective at treating chronic pain.

The researchers found there was “reliable evidence” that duloxetine offers effective pain relief, but also concluded that standard doses are as effective as higher doses. It was “consistently the highest rated medication” and was found to be equally effective at treating fibromyalgia, musculoskeletal and neuropathic pain conditions.

Other results showed that milnacipran can also reduce pain, but because it was included in fewer studies than duloxetine the evidence was less conclusive.   

Lead author professor Tamar Pincus from the University of Southampton said: “This is a global health concern. Chronic pain is a problem for millions who are prescribed antidepressants without sufficient scientific proof they help, nor an understanding of the long-term impact on health.

“Our review found no reliable evidence for the long-term efficacy of any antidepressant, and no reliable evidence for their safety for chronic pain at any point. Though we did find that duloxetine provided short-term pain relief for patients we studied, we remain concerned about its possible long-term harm due to the gaps in current evidence.

“We simply cannot tell about other antidepressants because sufficiently good studies are not available – but it does mean that people should stop taking prescribed medication without consulting their GP.”

University of Southampton researcher Dr Hollie Birkinshaw said: “Though previous investigations show that some antidepressants might relieve pain, there has never been a comprehensive study examining all medications across all chronic conditions – until now.

“Adopting a person-centred approach is critical to treatment, and when patients and clinicians decide together to try antidepressants, they should start from the drug for which there is good evidence.”

Statistician and review co-author Gavin Stewart from Newcastle University said: “We are calling on governing health bodies NICE and the FDA to update their guidelines to reflect the new scientific evidence, and on funders to stop supporting small and flawed trials. Evidence synthesis is often complex and nuanced but the evidence underpinning the use of these treatments is not equivalent, so current treatment modalities are hard to justify.”