How does paracetamol compare with placebo for adults with knee or hip osteoarthritis?

Abstract

Compared with placebo for adults with knee or hip osteoarthritis, paracetamol 1950 to 4000 mg per day, given in three to four divided doses, provides little or no improvement in pain (mean difference, 3.23 to 3.32 points) and function (mean difference, 1.83 to 2.92 points) on 0 to 100 scales (high‐ to moderate‐certainty evidence). There was no statistically or clinically significant difference between groups in terms of walking disability (mean difference, 0.55 to 0.06 on a 50‐foot walking test) or patient adherence (721 vs 759 per 1000 people). Fewer people used rescue medication with paracetamol than with placebo, but in one extreme of the 95% confidence interval, this difference could be unimportant. Moderate‐ to high‐certainty evidence shows little or no difference between groups in withdrawals due to adverse events (92 vs 78 per 1000 people) or overall adverse events (265 vs 263 per 1000 people), but more people experienced abnormal liver function when taking paracetamol than when taking placebo (46 vs 12 per 1000 people).