Osteoarthritis of the hip or knee is a highly prevalent condition associated with pain, stiffness, and functional impairment, eventually affecting the patient’s functioning and quality of life. Pain actually is the most common and burdensome symptom of osteoarthritis, and pain management is a pillar of its treatment.
Paracetamol is a medication that inhibits the production of prostaglandins, which are mediators of fever, pain, and inflammation. Although providing a weaker anti-inflammatory action than other medications, such as the non-steroid anti-inflammatory drugs (NSAIDs), it also is safer, especially for those at risk of gastrointestinal ulcer, and this is why current guidelines recommend paracetamol as a first-line analgesic for hip or knee osteoarthritis.
The question, however, is whether paracetamol provides clinically relevant benefits in the management of pain related to osteoarthritis of the hip or knee, or, on the contrary, its effects are comparable to those of a placebo (sugar pills or similar fake medication). Indeed, recent studies comparing paracetamol vs placebo failed to find additional benefits of paracetamol in improving pain, function, or quality of life for patients with hip or knee osteoarthritis. 
This issue was thus addressed by a recently published Cochrane Review by Leopoldino et al., who systematically reviewed published evidence, identifying ten high quality trials that compared paracetamol vs placebo, involving 3,541 participants with hip or knee osteoarthritis. In these studies, the paracetamol dose varied from 1.95 g/day to 4 g/day, Harms and benefits of either intervention were assessed by following up with patients for 3 weeks up to 3 months, 
Pooled results revealed that, independent of the administered dose, paracetamol was associated with only minimal improvements in pain and function in hip or knee osteoarthritis and with an overall increase in risk of adverse events. Paracetamol was found to be associated with a higher rate of serious adverse events, withdrawals due to adverse events, and a higher rate of abnormal liver function tests, but the small numbers of cases reported prevented conclusions on whether paracetamol might significantly increase such risks.
Results of this review have a highly relevant clinical impact and call for urgent reconsidering of current guidelines as to the recommendation of paracetamol for hip and knee osteoarthritis. Indeed, these results provide further support for refocusing pain management recommendations in hip and knee OA, from pharmacological treatments to lifestyle interventions, such as physical activity programs and control of obesity, whose benefits are supported by high quality, uncontroversial evidence.

Comment by Francesca Cecchi