Spiro 2006.

Study characteristics
Methods	Placebo and randomised controlled trial over 12 months
Participants	283 children aged 6 months to 12 years were recruited in an emergency department in Connecticut, USA. 138 of these 283 children were randomised to delayed antibiotics. Age: for the 138 children randomised to delayed antibiotics, the average age was 3.6 years; for the 145 children randomised to immediate antibiotics, it was 3.2 years Sex:delayed antibiotics group: 79 males, 59 females; immediate antibiotics group: 76 males, 69 females Exclusion criteria for this study included intercurrent bacterial infection, toxic appearance of child, patient hospitalisation, immunocompromise, child had been treated with antibiotics in the preceding 7 days, myringotomy tubes, current tympanic membrane perforation, uncertain medical access, uncertain telephone access, primary language of guardian other than English or Spanish
Interventions	Children were randomised to delayed antibiotics (advised to delay for 48 hours and the script was to expire after 72 hours) or immediate antibiotics. The clinician chose the antibiotic.
Outcomes	Primary outcome measures: fever, duration of fever, pain, duration of pain, antibiotic use Secondary outcome measures: adverse effects of antibiotics including vomiting, diarrhoea and rash Outcomes were measured by telephone interview by a research assistant with the caregivers of the included children
Notes	This study was supported by funding from a grant from the US National Institutes of Health, a grant from the Yale University School of Medicine and material support from Friends of Yale‐New Haven Children's Hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐assisted randomisation
Allocation concealment (selection bias)	Low risk	Sealed, opaque envelopes
Blinding (performance bias and detection bias) All outcomes	High risk	Study participants were not blinded, but outcome assessors were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Of the 138 participants randomised to delayed antibiotics, outcome data were reported for 132 participants. Of the 145 participants randomised to immediate antibiotics, outcome data were reported for 133 participants. ITT analysis was conducted.
Selective reporting (reporting bias)	Low risk	Prespecified outcomes were reported.
Other bias	Low risk	Funded by government body

AOM: acute otitis media
GABHS: group A beta‐haemolytic streptococcus
IQR: interquartile range
ITT: intention‐to‐treat
SD: standard deviation