In a new study on pain-relief medication for osteoarthritis, researchers found that paracetamol does not meet the minimum standard of clinical effectiveness in reducing pain or improving physical function in patients with knee and hip osteoarthritis.
Paracetamol No Good for OA

As indicated in the news release, “The study, published in The Lancet, is the largest analysis of randomized trials of medical pain relief for osteoarthritis to date, and finds that diclofenac 150mg/day, a non-steroidal anti-inflammatory drug (NSAID), is the most effective short-term pain relief. However, the authors caution against long-term use of NSAIDs because of known side-effects.”
“Dr. Sven Trelle from the University of Bern, Bern, Switzerland, and colleagues, pooled data from 74 randomised trials published between 1980 and 2015. With data from a total of 58, 556 patients with osteoarthritis, the study (a network meta-analysis) compared the effect of 22 different medical treatments and placebo on pain intensity and physical activity. The 22 treatments included various doses of paracetamol and seven different NSAIDs.”
“Paracetamol did not reach the minimum clinically important difference. In comparison, diclofenac at the maximum daily dose of 150 mg/day was most effective for the treatment of pain and physical disability in osteoarthritis (effect size -0.57), and superior to the maximum doses of frequently used NSAIDs, including ibuprofen, naproxen, and celecoxib.”
Dr. Trelle told OTW, “Paracetamol has no clinically relevant effect in this patient population and that there is no such thing as a class effect of NSAIDs but rather different effectiveness among different preparations namely that diclofenac and etoricoxib are the most effective ones.”
“Do not use single-agent paracetamol. Discuss with patients different NSAID options and weigh safety issues (especially GI and cardiovascular) with effectiveness.”

Discussion
This is a fascinating development. In my practice we've seen similar outcomes with the revised protocol. The key differentiator seems to be patient selection criteria. Has anyone else noticed the correlation with BMI thresholds?
Great point. I'd push back slightly on the conclusion, the sample size in the cited study is too small to draw population-level inferences. That said, the directional signal is compelling and worth a larger RCT.
We implemented a similar approach last year. Early results are promising but we're still gathering 12-month follow-up data. Happy to share our protocol if anyone is interested.
Join the conversation
Orthopedic professionals are discussing this. Sign in and upgrade to read every comment and add your voice.