These 2 studies are remarkable in several ways. First, both used large numbers of children who met stringent criteria for entry. Both sets of investigators feel that the stringent entry criteria accounted for the high rates of failure in the placebo groups – failure rates that were much higher than previously found. It stands to reason that if previous trials had less rigorous criteria for diagnosis (thereby including more children with nonbacterial middle ear disease) then the failure rate in those who are not treated would be lower in those trials.
The benefits to the patients seem to be consistent between the 2 studies, but one study also demonstrated reduced parental time off from work. As noted in an accompanying editorial, the failure rates of observation-only groups (33%-51% in these studies) lead to questions about whether observation alone is adequate, especially in children with severe findings.
In fact, Tähtinen and colleagues comment that the failure rate in their placebo group was much higher than in previous meta-analyses and raise the point that many previous studies used less stringent diagnostic criteria at entry. Therefore, if less stringent criteria for diagnosing AOM were used in past trials, fewer of those children would have experienced clinical failure given that they might not have had AOM in the first place. No matter how one views the issue of treating or not initially treating AOM, these 2 studies provide some of the best data yet of what to expect with either observation or active treatment of AOM in young children.