Study	Reason for exclusion
Batieha 2002	185 participants with acute respiratory tract infection from 2 health centres in Jordan. Assignment to antibiotic (azithromycin) was by means of alternation, not randomisation. At follow‐up of 3 days, 1 week, and 2 weeks, participants administered azithromycin or placebo did similarly in terms of the proportions improved or cured and duration of illness. The authors of the study concluded that routine use of antibiotics (azithromycin) in acute respiratory tract infection is unlikely to alter the course of the illness.
Christ‐Crain 2004	Randomised controlled trial concerned with application of a diagnostic test (serum calcitonin precursor, procalcitonin) that is raised in bacterial infections. 243 people admitted to hospital with suspected lower respiratory tract infections were randomly assigned to standard care (n = 119) or procalcitonin‐guided treatment (n = 124). On the basis of serum procalcitonin concentrations, use of antibiotics was more or less discouraged (< 0.1 µg/L or < 0.25 µg/L) or encouraged (≥ 0.5 µg/L or ≥ 0.25 µg/L), respectively. Re‐evaluation was possible after 6 to 24 hours in both groups. Primary endpoint was use of antibiotics. 59 (24%) participants had diagnosis of "acute bronchitis". Antibiotic use decreased in the procalcitonin group. Withholding antibiotic treatment based on procalcitonin measurement did not compromise patient outcome.
Dowell 2001	Randomised controlled trial of "delayed" versus "immediate" antibiotics for acute cough. Participants randomised to "delayed" arm were asked to wait a week before collecting their prescription. 55% of participants did not pick up their prescription. More participants were satisfied and "enabled" in the immediate‐treatment arm.
Gordon 1974	Participants were children with "symptoms referable to the respiratory tract", therefore likely many had upper respiratory infections (78% to 96% had runny nose, 74% to 83% had inflamed nasal mucosa).
Gottfarb 1994	Post‐randomisation exclusion of 23% of the sample due to laboratory evidence of pertussis infection. Outcomes not clearly reported.
Stephenson 1989	Participants were adults with upper respiratory infection. Not all had cough, and no information available on the subgroup of patients with productive cough.
Thomas 1978	Explicit data from the study were not published and the data are no longer available.