A population-based study linked a registry of 5,202 children diagnosed with type 1 diabetes under the age of 16 in New South Wales, Australia between 1990 and 2009 to hospital, emergency department, public insurance and death records covering 2001 to 2019. Age at onset was grouped as under 7 years (1,694 children), 7 to under 13 (2,538) and 13 to under 16 (970). Hazard ratios were adjusted for year of diagnosis, sociodemographic factors and acute diabetes complications.
Within 20 years of onset, complication rates ranged from 9.2 per 10,000 person-years for cardiac complications to 53.4 per 10,000 for kidney complications. Compared with diagnosis at 13 to under 16 years, diagnosis before age 7 carried lower risk of almost every complication: peripheral neuropathy adjusted hazard ratio 0.16 (95% CI 0.08 to 0.32), severe retinopathy 0.25 (0.17 to 0.36), cardiac complications 0.32 (0.15 to 0.66), kidney complications 0.40 (0.30 to 0.53) and lower-limb vascular complications 0.41 (0.24 to 0.69). Lower-limb infection was the single exception. Risk did not differ by attained age.
This runs against the intuition that earlier onset means longer exposure and therefore more damage. It suggests the prepubertal years are relatively protected and that the complication clock effectively starts running at puberty. For a clinic seeing young adults, that reorders who needs watching most closely.
At the first adult clinic visit, record age at diagnosis and treat it as a risk modifier rather than a piece of history. Someone diagnosed at 14 and now 30 warrants closer complication surveillance than someone diagnosed at 5 and also 30, despite the shorter duration. This is an observational registry study in a high-income, largely non-Asian population, so the direction is likely to hold but the absolute rates will not transfer to Indian practice, where access to continuous glucose monitoring (CGM) and pump therapy through childhood is far more variable.
- Record age at diagnosis prominently in the adult record, not just duration of diabetes.
- Prioritise retinal screening and albuminuria testing for those diagnosed in adolescence.
- Do not reassure on short duration alone if onset was post-pubertal.
- Foot checks apply regardless of onset age: lower-limb infection was the one complication that did not follow the pattern.
- Kidney complications were the commonest outcome at 53.4 per 10,000 person-years, which makes albuminuria testing the highest-yield screen in this group.
The statistics, in plain English
A hazard ratio of 0.16 for peripheral neuropathy means roughly a sixth the rate, but the interval from 0.08 to 0.32 is wide, so the true effect could be anywhere between a twelfth and a third. The direction is certain, the size is not. Because none of these intervals crosses 1.0, each difference is statistically real rather than a chance finding. This is a registry linkage and not a trial: children diagnosed before age 7 were also diagnosed in different calendar years and cared for differently, and adjusting for year of diagnosis reduces but cannot remove that. The finding that risk did not differ by attained age means the pattern held whether you counted time from diagnosis or from birth, which strengthens it.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for diabetes & endocrinology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free