• wounds and injuries
• wound infection
• soft tissue infections

In their Emergency Medicine Journal paper, Zwaans et al present a randomised controlled trial that asked whether the use of non-sterile gloves and dressings affects risk of infection when suturing traumatic wounds in the ED.1

Most randomised trials compare two treatments so they can determine which is better. However, such a ‘superiority’ design would not make sense when comparing the effect of sterile and non-sterile gloves as there is no reason to think that non-sterile gloves should be superior in terms of reducing wound infections. There may nevertheless be other reasons to prefer non-sterile gloves if they are ‘not much worse’—perhaps because they are cheaper, more accessible or generate less packaging waste. Zwaans et al therefore chose a non-inferiority trial design. Instead of asking whether sterile or non-sterile gloves are ‘better’, their trial asked whether non-sterile gloves are ‘much worse’ than sterile equivalents.

To answer such a question, Zwaans et al first had to specify ‘how much worse’ they would be willing to accept. This ‘non-inferiority margin’ is a matter of judgement and depends on factors such as the downsides of sterile gloves (eg, cost, accessibility and environmental burden) and the magnitude of the risk they seek to avoid (eg, wound infection). The trial team chose a non-inferiority margin of 2%, which would have allowed them to conclude that non-sterile gloves are ‘non-inferior’ if the upper bound of the 95% …