Earlier egg introduction shows up in the population, not just the trial

Design

Two population-based cross-sectional samples recruited by identical methods, with regression standardisation for risk factors

Population

7,209 infants aged 11-15 months in Melbourne, Australia: 5,276 in 2007-2011 and 1,933 in 2018-2019

Primary outcome

Challenge-proven egg allergy after skin prick testing

Effect

Prevalence 9.2% to 7.6%, adjusted absolute difference -1.6 percentage points (95% CI -3.3 to -0.005); in infants with early eczema 34.6% to 21.9%, difference -12.7 (-20.0 to -5.4)

Randomised trials showed that introducing egg earlier reduces egg allergy, and guidelines worldwide changed accordingly. Whether that translated into fewer allergic children is a different question, and one that trials cannot answer.

Two Melbourne cohorts were recruited by identical methods at the 12-month immunisation visit, one before the guideline change (2007-2011, 5,276 infants) and one after (2018-2019, 1,933). All had skin prick testing, and positive results went to oral food challenge — so the outcome is challenge-proven allergy, not parental report. Median age at egg introduction fell from 8 months to 6.

After standardising for known allergy risk factors, egg allergy prevalence fell from 9.2% to 7.6%, an adjusted absolute difference of 1.6 percentage points (95% CI -3.3 to -0.005). Among infants with early eczema — the group at highest risk and the group where clinicians most often hesitate — it fell from 34.6% to 21.9%, a difference of 12.7 percentage points (-20.0 to -5.4).

That second figure is the one to carry into clinic. The advice to introduce egg early is hardest to give to the parent of an infant with eczema, and that is exactly where the population effect is largest.

• Advise egg introduction around 6 months, alongside other solids
• Do not defer introduction in an infant with eczema — that is where the gain is largest
• Treat eczema actively at the same time rather than using it as a reason to wait
• Outcomes here were challenge-proven, so the prevalence figures are not inflated by parental report
• Once introduced, egg needs to stay in the diet regularly

Why it matters

It closes the loop between a trial result and what actually happens to children when advice changes.

Don't overread it

Two cross-sectional samples a decade apart cannot prove the guideline caused the fall, only that prevalence fell alongside it.

The statistics, in plain English

The whole-population difference of 1.6 percentage points has a confidence interval whose upper limit is -0.005, which just clears zero: statistically significant, but only just, and the true effect could be very small. The eczema subgroup result is both larger and more robustly separated from zero (-20.0 to -5.4). This is a before-and-after comparison of two cross-sectional samples, adjusted for measured risk factors, so anything else that changed in Melbourne over that decade is a potential explanation.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.