Abstract
OBJECTIVE: To explore the effect of a pneumococcal conjugate vaccine on the risk of otitis media with effusion and to search for subgroups in which the vaccine had a higher or lower effect. METHODS: Analyses were performed on data from the Finnish Otitis Media Vaccine Trial, a randomised controlled double-blind trial to evaluate the efficacy of pneumococcal conjugate vaccination against acute otitis media. Data on the vaccination effect against otitis media with effusion were obtained by means of symptom interview and pneumatic otoscopy during pre-scheduled follow-up visits at the age of 7 and 24 months. Two endpoint definitions were considered: otitis media/tube (otitis media or tympanostomy tube in situ (OM/T)) as the primary endpoint and otitis media with effusion as the secondary endpoint. No evidence was found of an age-dependent association with vaccination effect. Therefore, the final marginal logistic regression analyses were performed on the combined data from the two follow-up visits. RESULTS: The risk of otitis media tended to be lower in the pneumococcal vaccine group. The odds ratio for otitis media/tube was 0.94 (95% confidence interval 0.77-1.14) and the odds ratio for otitis media with effusion was 0.90 (95% confidence interval 0.69-1.19). Presence of older siblings increased the risk of otitis media/tube and otitis media with effusion at 7 months of age. In addition, it appeared that children without older siblings and attending day-care at 24 months of age tended to benefit more from the pneumococcal conjugate vaccine. In this subgroup, the odds ratio for otitis media/tube was 0.81 (95% confidence interval 0.55-1.20) and for otitis media with effusion the odds ratio was 0.43 (95% confidence interval 0.22-0.86). CONCLUSION: The effect of pneumococcal conjugate vaccination on the risk of otitis media with effusion was concordant with the efficacy seen against acute otitis media, although not distinguishable from no effect in the overall analysis. In children without older siblings, vaccination appeared to reduce the point prevalence of otitis media with effusion; this effect was not apparent in children with older siblings.
