Little 1997.

Study characteristics
Methods	Open randomised controlled trial over 20 months
Participants	712 adults and children aged ≥ 4 years with sore throat presenting to 11 general practices in England, UK. Of these 712 participants, 235 were randomised to delayed antibiotics. Age: of the 235 participants randomised to delayed antibiotics, 181 were older than 12 years; of the 246 participants randomised to immediate antibiotics, 187 were older than 12 years; and of the 232 participants randomised to no antibiotics, 173 were older than 12 years Sex:delayed antibiotics group: 82 males, 153 females; immediate antibiotics group: 95 males, 151 females; no antibiotics group: 82 males, 150 females Exclusion criteria: people were excluded if they had a sore throat that was clearly not a bacterial infection, e.g. due to drugs, aphthous ulcers, candidal infection. Other exclusion criteria included being very unwell, suspected or previous rheumatic fever, multiple (more than 5 per year) attacks of tonsillitis, quinsy and pregnancy.
Interventions	Participants in the delayed antibiotics group were instructed to pick up a script left at reception after 72 hours if needed. Participants in the immediate antibiotics group were immediately offered a script for antibiotics. The antibiotic prescription for both groups was penicillin V 250 mg 4 times a day for 10 days. For children aged 3 to 5 years, the dose was reduced to 125 mg. Participants who were penicillin allergic received a script for erythromycin with the same dosing regimen as for penicillin. Participants in the no antibiotics group were not offered antibiotics.
Outcomes	Primary outcomes: fever, cough, duration of pain and duration of malaise. Antibiotic use and patient satisfaction were measured. Secondary outcomes: absences from school, diarrhoea, stomach ache, rash Outcomes were assessed using a patient diary and a follow‐up telephone call from a research assistant
Notes	This study was supported by Wessex NHS regional research and development funds. We approached the authors for study data, which they provided.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomisation method not described
Allocation concealment (selection bias)	Unclear risk	"Sealed envelopes", but no mention of opacity
Blinding (performance bias and detection bias) All outcomes	High risk	This study was described as an open randomised trial, so no blinding was used
Incomplete outcome data (attrition bias) All outcomes	Low risk	Intention‐to‐treat analysis conducted. In the delayed antibiotic group, 179 participants responded out of 235. In the immediate antibiotic group, 215 participants responded out of 246. In the no antibiotic group, 186 participants responded out of 231.
Selective reporting (reporting bias)	Low risk	Outcomes were reported as indicated in the methods section
Other bias	Low risk	Funded by government body