Honey vs. pharmacological antitussives for acute cough in children: meta-analysis finds comparable outcomes, with honey linked to better child sleep

The primary analysis found no statistically significant differences between honey and medications for most cough outcomes (frequency, severity, bothersomeness) over a 3-day follow-up. However, one key finding emerged: cough's effect on the child's own sleep was significantly reduced in the honey group, with a mean difference of −0.73 points on the measured scale (95% CI: −1.45 to −0.01; p = 0.05). A post-hoc sensitivity analysis suggested that honey may perform better than pharmacological agents across all outcomes, including caregiver sleep disruption and global symptom scores, although the authors note these exploratory results are hypothesis-generating rather than conclusive. Subgroup analyses showed consistent patterns regardless of which specific medication was used as the comparison. A major limitation is the short follow-up duration (3 days), which may not capture longer-term patterns or relapse. The authors conclude that given the established risks of pediatric antitussives and honey's safety profile and low cost, honey warrants promotion as an alternative approach.

The primary pre-planned analysis found no statistically significant advantage for honey on cough frequency, severity, or bothersomeness—the effect sizes overlapped zero and the heterogeneity between studies was high, suggesting variable study quality or population differences. The child sleep finding (MD −0.73) survived the primary threshold but just barely (p = 0.05), indicating a modest and borderline effect. The exploratory post-hoc sensitivity analysis—which removed one study at a time and re-ran the analysis—showed much larger benefits for honey across all domains, but this approach inflates the risk of false positives and should be interpreted cautiously. Among the honey-vs-dextromethorphan subgroup specifically, benefits emerged for cough frequency and severity, suggesting the effect may depend on the medication class. Follow-up was limited to 3 days, too short to assess whether symptom relief persists, whether cough recurs, or whether sleep quality rebounds. The mean age of 3–5 years is narrow; effects in infants or older children remain unclear. The authors emphasize that their hypothesis about honey superiority rests on post-hoc exploration, not the pre-specified analysis plan.