The review included 20 studies (based on supplementary online table S3) comprising 14 RCTs, two cluster randomised trials, three pre/post-test studies and one non randomised controlled trial.
Interventions to influence parental consulting
Eight studies reported on interventions aimed to influence parental consulting (1,488 parents, 1,580 families, 558 children) with follow-up that ranged from three days to 17 months. There was likely to be risk of bias for all but two of the studies, which were judged to be at minimal risk.
Five studies assessed change in parent knowledge relating to consulting. All five studies reported significant increases in knowledge about respiratory tract infections or appropriate reasons to consult in the intervention groups compared with control groups. None of studies measured the impact on the actual number of consultations.
Six studies reported on change in consultation rates. Three studies found providing booklets before the children became ill reduced the number of consultations, but two of these studies were published over 20 years ago. Two studies assessed interventions delivered at the point of consultation which aimed to reduce re-consultations; neither of the studies found any difference in consultation rate, although one study reported that the intervention group was more likely to receive antibiotics.
Interventions to influence parents' decisions in antibiotic use for respiratory tract infections in children
Nine studies focused on this topic (2,916 participants) with follow-up that ranged from one day to 36 weeks. Most interventions took place during the consultation.
Four out of five studies measuring parental knowledge of appropriate antibiotic use found a significant increase following the intervention compared to the control group. Three studies were rated as likely to be at risk of bias, one study was rated as minimal and one study was highly likely to be biased.
Two studies measured parental attitudes towards antibiotic use and reported mixed results. Neither of the studies found any change in antibiotic prescribing rates over six to 12 months. One study was judged to be at minimal risk of bias; the other study was given a high rating.
Two studies assessed parent satisfaction with a watchful-waiting approach but no differences were found between intervention and control groups. Both studies were rated at minimal risk of bias.
Interventions to influence antibiotic use in children with respiratory tract infections
Five studies reported this outcome. All but one study assessed the effect of delayed prescribing or watchful waiting; the other study used a book-based intervention. Meta-analysis of five studies (all rated minimal risk of bias) found a significant reduction in numbers of children taking or parents filling antibiotic prescriptions (RR 0.39, 95% CI 0.29 to 0.53), but significant heterogeneity was observed (Ι²=77%). Analysis with one study (where all parents used antibiotics) removed still found a significant effect of the intervention (RR 0.46 95% CI: 0.40 to 0.54).