Aims To assess, using Cochrane methodology, the current pharmacological treatments for gastro-oesophageal reflux disease (GORD) for the Cochrane Upper GI&Pancreatic Diseases group.

Methods We searched for RCTs in the Cochrane Central Register of Controlled trials, MEDLINE, EMBASE, CISCOM and ISI Science Citation Index, ISI web of science and Meta-Register of Controlled Trials. Handsearching of references was performed including published abstracts from gastroenterology conferences over 5 years. Abstracts were reviewed and relevant RCTs selected for all children with functional reflux/GORD (birth–16 years) receiving medications. Trials were critically appraised, data independently extracted and risk of bias assessed using RevMan. Strength of evidence is expressed using GRADE methodology.

Results We considered the efficacy of medications in treating GORD, and functional reflux, in infants, and GORD in older children. Twenty-four studies (1201 patients) contributed data.

There is good evidence proton-pump inhibitors (PPIs) in children and infants with GORD, but benefit is unclear for infants with functional reflux. Two similar studies of infants with GORD were meta-analysed, assessing PPIs against placebo/gaviscon. PPI was not significantly superior in infants at reducing reflux-index (SMD–0.28, 95% CI –1.98 to 1.42).

Moderate evidence exists for H2-antagonist efficacy in improving erosive oesophagitis; meta-analysis was not possible.

Insufficient evidence exists to judge efficacy of Domperidone. Evidence of Gaviscon Infant’s efficacy in infants is limited by small numbers and outcomes.

Conclusions Despite the paucity of studies, there is good evidence supporting the use of PPIs in older children and infants with GORD, and moderate evidence supporting H2-antagonists. Concerns regarding pharmaceutical-company support, and study heterogeneity are discussed. A lack of independent placebo-controlled or head-to-head trials with common end-points remains.

For infants with functional reflux, poor correlation between symptomatology and investigative findings limits assessment of treatment efficacy. PPIs have some evidence of efficacy, with significant study heterogeneity. Domperidone has poor evidence of efficacy. Gaviscon Infant has some evidence to support use, limited by short follow-up. The lack of robust RCT evidence for treatment for preterm-babies or children with neurodisabilities is discussed. We welcome the opportunity to present the findings in detail. A template for further research is outlined.