IT IS the world’s most commonly used analgesic, but good evidence for the effectiveness of paracetamol only exists for four conditions, and even then, the benefits were small, a large review has found.

Published in the MJA, an analysis of 36 systematic reviews of randomised controlled trials identified just four conditions for which there was high or moderate quality evidence that paracetamol relieved pain – knee or hip osteoarthritis, craniotomy, tension-type headaches, and perineal pain after childbirth.

However, the size of the pain reduction was so small in the case of osteoarthritis it was “arguably clinically insignificant”, lead author, pharmacist Dr Christina Abdel Shaheed of the University of Sydney told InSight+.

Paracetamol provided mean pain relief of just 0.3 points on a 0–10 point scale for patients with knee and hip osteoarthritis who were given 1 g four times daily.

Furthermore, paracetamol was of no benefit in relieving acute low back pain, the review found, citing a 2016 Cochrane review.

For all other conditions, there was insufficient evidence that paracetamol was effective.

The authors concluded:

“Evidence for the efficacy of paracetamol in most pain conditions is of low quality or inconclusive, and for the four conditions for which there is high or moderate quality evidence of efficacy, the benefits are small.”

In people with tension-type headaches, a single dose of up to 1 g paracetamol increased the likelihood of being pain-free at 2 hours by 30% compared with placebo (risk ratio [RR], 1.3; 95% CI, 1.1–1.4).

Among women with early post partum perineal pain, those who received 1 g paracetamol were more likely to achieve 50% pain relief compared with those given placebo (RR, 2.4; 95% CI, 1.5–3.8). Lower doses were also effective.

Most studies in the review evaluated the immediate effects of single doses, unlike typical clinical practice. The authors said more high quality studies were needed, especially into the efficacy and safety of long-term use of paracetamol.

“The clinical application of paracetamol is primarily guided by low quality evidence, at best,” the authors concluded.

Dr Abdel Shaheed expressed concern that many patients with musculoskeletal conditions were over-reliant on high strength paracetamol formulations.

“Some patients are convinced they get relief with long-acting high strength formulations of paracetamol, but for conditions like acute low back pain, the reduction in pain is most likely due to the natural course of the illness,” she told InSight+.

Dr Abdel Shaheed said patients with osteoarthritis who felt they experienced a benefit from paracetamol could continue to use it according to the label; “however, they should be aware that the benefits are likely to be small”.

“If they want to optimise pain relief, they should try to combine it with other non-pharmacological strategies such as exercise, heat-wraps and physiotherapy,” she added.

Although paracetamol was one of the safest analgesics when taken at recommended doses, Dr Abdel Shaheed said overdosing was a relatively common problem – particularly with the long-acting formulations.

“It is very easy to accidentally overdose on paracetamol, either by taking too much of a long-acting preparation, or by adding other medicines which also contain paracetamol, such as certain cold and flu tablets,” she said.

“Unfortunately, even a 1 g overdose of paracetamol – two tablets above the maximum – has landed people in hospital with paracetamol toxicity.”

The latest review found the frequency of adverse events with paracetamol was similar to that with placebo, but the authors stressed this finding was limited given most studies involved only single doses.

Professor Ric Day, a pharmacologist and toxicologist at the University of New South Wales, co-authored the latest review and was also an investigator on a 2014 trial that led to Cochrane’s recommendation against paracetamol for acute low back pain.

“Paracetamol is an analgesic so there’s no doubt you’re going to get some relief with it from simple things like headaches and toothaches. The question is how much?” Professor Day told InSight+.

“The evidence clearly shows paracetamol is of no clinically relevant benefit on average in acute low back pain and the guidelines are starting to change to reflect that.

“But when people come in with osteoarthritis and you are going through what you have to offer them by way of pain relief, it’s still reasonable to say to them ‘paracetamol might help’. It may only be a placebo effect, but is that a bad thing if they get some benefit?”

“We’re pushing hard for evidence-based non-drug approaches, but not every patient is ready and willing to leap right into those things.”

Professor Rachelle Buchbinder, a rheumatologist and coordinating editor of Cochrane Musculoskeletal, who was not involved in this MJA study, told InSight+:

“There is a paucity of evidence for use of paracetamol for many conditions that it is used for. More trials like those that have been done for acute low back pain are needed to get definitive answers.

“It would also be worthwhile to consider the types of pain that paracetamol might be more effective for – are there some pain mechanisms for which it might be effective and others where there might not be a strong rationale?”

For acute low back pain, Professor Buchbinder said most cases improved irrespective of treatment. Exercise therapies provided some benefit on a population level, she added.

“Good evidence-based advice and information and addressing misconceptions is most important,” she said.

Professor Buchbinder shares research grants with several authors of the latest review.

Also online first at the MJA

Podcast: Professor Henry Woo, Director of Uro-Oncology at the Chris O’Brien Lifehouse, and Dr Amy Teh, a radiation oncologist at Icon Cancer Centre, talk about treatments for prostate cancerOPEN ACCESS permanently.

Research: Factors that influence whether patients with acute coronary syndromes undergo cardiac catheterisation
Ayad et al; doi: 10.5694/mja2.50997 … OPEN ACCESS permanently.

Research letter: Changes in the proportions of authors in Australian medical journals who were women, 2005–2018
Lennon et al; doi: 10.5694/mja2.50998 … OPEN ACCESS permanently.



5 thoughts on “Paracetamol for osteoarthritis pain “arguably clinically insignificant”

  1. Anonymous says:

    Whilst a statistician may be able to show that the effect of paracetemol is not worth the effort, there are many patients , including me, with musculoskeletal pain who definitely benefit by taking even a modest dose.
    A recent injury left me with back/sacro iliac pain which has inhibited many activities. I have only rarely taken analgesics in the past and do not take NSAIDs, yet one paracetemol allows me to increase my activity with less pain for about four hours. It could be argued that it is a placebo effect but I am reluctant to take any tablet so not really prone to placebo effect. Papers such as this are taken up by the media and subsequently the patients who will now tell me that paracetemol is ineffective before they even start taking it and demand something stronger such as an opiate. Even it there was a placebo effect for many, that in itself is beneficial and may stop patients starting on addictive drugs. I do not think such papers advance medical practice. Whilst it is good to question, I do not think this paper, which is not original research, has improved our ability to manage pain.

  2. Anonymous says:

    I am sorry, but this is just a ‘beat-up’ by people pushing their own agenda’s and lead to more confusing media information to the general public.


    That is, just because this most current study did not show benefit for, say acute low back pain, it has not proven that paracetamol is ineffective for all acute back pain.

    There are many confounding factors: Most likely the study was not powered enough with sufficient ‘study individuals’ to determine no benefit or harm from the intervention; It has long been know that there is a proportion of the population who do not respond to paracetamol because of genetic differences in their paracetamol pharmacokinetics.

    Our “medical toolbox” to treat pain is already very limited. This sort of mass media misinterpretation of results is really very dangerous if all medical professionals, or our patients, suddenly decide not to use paracetamol!

  3. Anonymous says:

    It is not an anti-inflammatory but does have some analgesic effect. Pharmacological Medicine 101 in 1973.
    Almost 50 years ago. Golly.

  4. Anonymous says:

    My experience treating patients with osteoarthritis of the hip and knee, albeit often with paracetamol 1330 mg tds, is that Paracetamol is effective as an analgesic agent for those patients. The Acute Pain Management Scientific Guidelines state “Paracetamol is an effective analgesic for acute pain; the incidence of adverse effects is
    comparable to placebo.”

  5. Anonymous says:

    I’m a bit sceptical about this finding, based on my experience as a doctor and personal use of Paracetamol. I think there is more to this finding and story that further studies will reveal.

    Post-operatively, on two occasions, I was prescribed a Paracetamol/Codeine compound but only took Paracetamol in fact and did fine. However, I am a slow acetylator, and stopped breathing after the first op on a normal dose of Pethidine post-operatively which lead to my caution about opioids. (It was an interesting experience in that I was conscious enough to turn the sats monitor around and take a few deep breaths when I saw it dropping.)

    These days I get by on one tablet of 500mg each time once or twice a day now for mild OA in both thumbs – attributable to past injury. (It also works for occasional mild dental pain.)

    I’m a bit of a sceptic generally and I don’t think I’m really very susceptible to placebo effects.

Leave a Reply

Your email address will not be published. Required fields are marked *