Design: systematic review and random-effects meta-analysis of randomised trials, PRISMA-C 2026, GRADE assessed
Population: 8 trials, 1,325 children aged under 18 with acute mild-to-moderate pain; 196 analysable for the primary outcome
Primary outcome: pooled pain score difference and proportion achieving pain freedom
Effect: standardised mean difference -0.28 (95% CI -0.57 to 0.00, p=0.052); pain freedom risk ratio 1.03 (0.53-1.99); low certainty

Eight randomised trials enrolling 1,325 children were identified in a review restricted to paediatric populations with acute mild-to-moderate pain. Three of them, with 196 analysable participants, contributed to the primary pooled pain score: standardised mean difference -0.28 (95% CI -0.57 to 0.00, p=0.052), a small effect favouring ibuprofen that did not reach significance. For pain freedom, two trials with 114 children gave a risk ratio of 1.03 (0.53 to 1.99). Adding an adult soft-tissue injury trial in a prespecified sensitivity analysis pulled the estimate further towards null, to -0.15 (-0.38 to 0.09).

Certainty was graded low for both primary outcomes, mainly for imprecision. Earlier meta-analyses that suggested an ibuprofen advantage had mixed adult and paediatric data; restricting to children removes it.

The practical reading is that the choice is a clinical one rather than an efficacy one. Where the pain is inflammatory - a fracture, an injury, dental pain - the mechanistic argument for ibuprofen remains, and the authors note the advantage may concentrate there. Where there is dehydration, vomiting, a bleeding risk, renal concern, asthma sensitive to NSAIDs, or a child too young for comfortable NSAID dosing, paracetamol is not the second-best option. Neither should be withheld while deciding.

Choose on contraindications and pain type, not on an assumed efficacy gap
Prefer ibuprofen where the pain is clearly inflammatory - fractures, injuries, dental pain
Prefer paracetamol where the child is dehydrated, vomiting, or has renal or bleeding concerns
Dose by weight and say the dose out loud to the parent - underdosing is the commoner failure than drug choice
Do not alternate the two reflexively; agree one agent, dose it properly, and reassess

Why it matters

It removes the assumed efficacy hierarchy between the two drugs every paediatric department uses most.

The statistics, in plain English

A p value of 0.052 with an interval running from -0.57 to exactly 0.00 is the textbook borderline result, and the honest reading is that the data are consistent with a small ibuprofen advantage and also with none. What settles the interpretation is the direction of the sensitivity analysis: adding more data pulled the estimate towards null rather than sharpening it. Note also how small the evidence base actually is - 1,325 children enrolled across eight trials, but only 196 contributed to the primary pooled outcome, which is why certainty was graded low for imprecision.