Before surgery, 3.7% of these children had normal hearing. After, 44.7% did

Australian Aboriginal and Torres Strait Islander children experience some of the highest rates of otitis media with tympanic membrane perforation in the world [s1]. A perforated eardrum in childhood is not a minor thing: it means recurrent infection, and it means hearing loss during the years a child is learning to speak and learning to read.

A retrospective cohort study published in the ANZ Journal of Surgery on 4 August reports ten years of surgical outcomes from Queensland's response to that problem [s1].

The programme

Deadly Ears is a statewide, Indigenous-led outreach service providing ear health care on Country across Queensland [s1]. The study evaluated myringoplasty — surgical repair of a perforated eardrum — delivered through the programme to Aboriginal and Torres Strait Islander children between January 2015 and February 2025 [s1].

Demographic, clinical, surgical and audiological data were extracted from medical records, with primary outcomes defined as tympanic membrane closure and improvement between pre- and post-operative audiological categories [s1].

One hundred and four children underwent 135 myringoplasties across nine communities, at a median age of 10 years (IQR 8–12) [s1].

The follow-up figure is the one to notice first

Surgical and audiological follow-up was complete for 123 of the cases — 91% — at a median follow-up of 707 days (IQR 358–1,156) [s1].

Ninety-one per cent complete follow-up at nearly two years, in remote and regional communities, is the number a reader should sit with. Loss to follow-up is the standard failure mode of outreach surgery; a programme that operates in a community and never learns whether the repair held has no way to know if it works. The authors attribute this directly to strong community engagement facilitated by Deadly Ears [s1], and it is the reason the rest of the results can be believed.

What the surgery achieved

Complete tympanic membrane closure was achieved in 60.2% of cases, and perforation size improved in 74.8% [s1].

Hearing improved by at least one audiological category in 64.2% of ears [s1]. And the proportion of children with normal hearing rose from 3.7% before surgery to 44.7% after [s1].

That last pair of numbers is the clinically meaningful one. Closure of the eardrum is the surgical endpoint; hearing is the endpoint that matters to a child in a classroom. Fewer than 4 in 100 of these children had normal hearing at the point they were referred for surgery — a figure that describes how severe the ear disease burden was before any operation was attempted.

What predicted a better result

Three factors were associated with improved tympanic membrane closure rates: cartilage-based grafts, a consultant surgeon operating, and shorter time to initial follow-up [s1]. Hearing improvement was associated with age under 12 [s1].

These are associations within a retrospective cohort, not randomised comparisons, and the usual caution applies — a consultant may take different cases than a trainee, and shorter time to follow-up may mark children whose families had easier access rather than any effect of the timing itself. The age finding is the most straightforwardly actionable: it points toward earlier surgery, though the study does not test an age threshold as an intervention.

The comparison the authors draw themselves

Their conclusion is candid on the point that matters most for policy. Surgical and audiological outcomes were consistent with other Aboriginal and Torres Strait Islander outreach programmes, but remain lower than metropolitan centres [s1].

A 60.2% closure rate is below what a well-resourced urban tertiary ear surgery service would expect from primary myringoplasty. The paper does not present that as a failure of the programme — it presents it alongside a 91% follow-up rate and a service delivered across nine communities where the alternative is generally no surgery at all. But it also does not hide it, and the persistence of the metropolitan gap is the finding that the programme's own data puts on the record.

Limits

This is a retrospective single-programme cohort with no control group. It cannot say what would have happened to these children without surgery, and it reports outcomes at a median of roughly two years, not into adulthood. The 104 children represent those who reached surgery through the programme, not all children in those communities with perforations — so the results describe outcomes of treatment, not the burden of untreated disease.

It also does not address the upstream question. Myringoplasty repairs damage that recurrent childhood ear infection has already done; nothing in this study speaks to preventing the infections.

What the authors ask for

Their stated conclusion is that ongoing investment in culturally responsive, community-designed outreach models is essential [s1].

The evidence they put behind that is specific rather than rhetorical: the 91% follow-up rate, attributed to community engagement [s1], is a measurable operational outcome of the model's design. An outreach programme that is Indigenous-led and delivered on Country produced follow-up completeness that outreach surgery frequently fails to achieve. That is the part of the result most transferable to other services.

What to watch

Whether the metropolitan outcome gap narrows as the programme accumulates cases, and whether the cartilage-graft and early-follow-up associations hold up prospectively. Neither is settled by a retrospective cohort of this size.

Sources

• [s1] Myringoplasty Outcomes for Aboriginal and Torres Strait Islander Children Treated Via the Deadly Ears Program: A 10-Year Retrospective Cohort Study, ANZ Journal of Surgery, published online 4 August 2026. https://doi.org/10.1111/ans.70871